Resonance-state selective photodissociation of OCS (2 1 ∑ +: rotational and vibrational distributions of CO fragments

Article abstract of DOI:10.1063/1.1310606

In spite of the long history of the pump-probe-type investigation of molecular dynamics using the laser light sources, the present study reports for the first time the pump-probe experiments in which two sets of tunable VUV laser light sources generated by a two-photon resonant four-wave mixing scheme are employed. The present study probes the rotational distribution of the CO fragments by measuring the laser-induced fluorescence (LIF) spectrum of the a ¹Φ-X ¹∑⁺ bands of CO using the second tunable VUV light source, and probe the S(¹S) fragments by tuning the UV light near the ³D^o - ¹S transition of the sulfur atoms to derive the vibrational distribution of the CO fragments through the measurements of the Doppler profiles. On the basis of the product state distributions, this paper discusses the mechanism which governs the later stage of the photodissociation of OCS after it leaves the Franck-Condor region of the 2 ¹∑⁺ surface.

Full text of DOI:10.1063/1.1310606

JOURNAL OF CHEMICAL PHYSICS
VOLUME 113, NUMBER 16
22 OCTOBER 2000
Resonance-state selective photodissociation of OCS „2 1⌺¿…: Rotational
and vibrational distributions of CO fragments
Ryuji Itakura,a) Akiyoshi Hishikawa, and Kaoru Yamanouchib)
Department of Chemistry, School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku,
Tokyo 113-0033, Japan
͑Received 6 June 2000; accepted 28 July 2000͒
The rotational and vibrational state distributions of the CO fragments produced through the
photodissociation of OCS in the vacuum ultraviolet ͑VUV͒ region ͑150–155 nm͒, OCS (2 1⌺ϩ)
→CO (X 1⌺ϩ)ϩS(1S), are derived for the three lowest quasi-bound vibrational resonances (
*
v
ϭ0Ϫ2) in the 2 1⌺ϩ state. The rotational state distributions of the CO fragments in the COϭ0 and
v
1 vibrational states are determined, respectively, by the analysis of the rotational structures in the
laser-induced fluorescence ͑LIF͒ spectra of the A1⌸–X 1⌺ϩ(0,0) and ͑1,1͒ transitions of CO. The
rotational temperatures of CO in the COϭ0 state are low ͑ϳ100 K͒ for all the three resonances,
v
*
while those in the
ϭ1 state are substantially higher, i.e., 2210, 940, and 810 K for ϭ0, 1, and
vCO
v
2, respectively. The vibrational state distributions of CO are derived from the Doppler spectroscopy
of the counterpart S(1S) fragments. From the analysis of the observed Doppler profiles, it is found
for all the three lowest vibrational resonances of OCS that the vibrational distributions are
represented well by the Boltzmann-type distribution with a vibrational temperature of around 7000
K. On the basis of these new findings, the energy partitioning in the photodissociation process
through these three vibrational resonances in the 2 1⌺ϩ state is discussed. © 2000 American
Institute of Physics. ͓S0021-9606͑00͒01740-2͔
I. INTRODUCTION
*
numbers, ϭ0–4, representing the quasi-bound vibrational
v
motion near the FC region on the potential energy surface
͑PES͒ of the 2 1⌺ϩ state, which is mostly repulsive along the
dissociation coordinate. This quasi-bound vibration was in-
terpreted as the in-phase CO and CS stretching mode perpen-
dicular to the dissociation coordinate. On the basis of the
PHOFEX spectrum exhibiting characteristic asymmetrical
peak profiles with different peak widths, the ultrafast nuclear
motion on the 2 1⌺ϩ PES near its FC region was discussed.
Even though we were able to understand the ultrafast
nuclear dynamics near the FC region from the absorption
features, there still remains a question of how the energy
imposed on molecules by the photoabsorption is partitioned
into the fragments in the dissociation process. When consid-
ering that the five main peaks, which form the progression of
the vibrational resonances, exhibit different characteristic
asymmetric profiles, it would be interesting to investigate
how the energy flow is influenced by the character of the
initially prepared resonant states.
In order to understand the energy partitioning dynamics,
it becomes necessary to probe the rotational and vibrational
distributions of the CO fragments. Since the first dipole-
allowed excited state, A1⌸, of CO is located in the energy
range of ϳ65 000 cmϪ1, we need to introduce another tun-
able VUV light source to probe the fragments while tuning
the first VUV light source to selectively excite the vibra-
tional resonance peaks in the 2 1⌺ϩ –1 1⌺ϩ transition of
OCS.
When molecules are irradiated by light in the vacuum
ultraviolet ͑VUV͒ wavelength region, they undergo photo-
dissociation in most cases. In the case of polyatomic mol-
ecules, which have more than or equal to three degrees of
vibrational freedom, the nuclear motion along the direction
perpendicular to the dissociation coordinate is often bound.
Therefore, even in the direct dissociation, they could form
quasi-bound vibrational resonances appearing as structured
profiles with broadened peaks in an absorption spectrum,
from which the nuclear motion near the Franck–Condon
͑FC͒ region could be extracted.1,2 However, until recently,
little effort has been made to interpret such rich spectral fea-
tures of polyatomic molecules in the VUV region. This may
be partly due to the overlap of the transitions from the rota-
tionally and vibrationally excited states, which broadens and
blurs originally characteristic spectral structures.
In our recent studies,3–7 the photodissociation process of
OCS, OCS (2 1⌺ϩ)→CO (X 1⌺ϩ)ϩS(1S), in the 140–160
nm wavelength region was studied by measuring the excita-
tion spectrum called the photofragment excitation
͑PHOFEX͒ spectrum under jet-cooled conditions by moni-
toring the S(1S) fragments employing a tunable VUV light
generated by the four-wave difference frequency mixing
technique. As shown in Fig. 1, the PHOFEX spectra of the
entire region of the 2 1⌺ϩ –1 1⌺ϩ band of OCS exhibited
five main peaks. These peaks were assigned to the transitions
to the resonant vibrational states with respective quantum
In the present study, we probe the rotational distribution
of the CO fragments by measuring the laser-induced fluores-
cence ͑LIF͒ spectrum of the A1⌸–X 1⌺ϩ bands of CO using
a͒
Research Fellow of the Japan Society for the Promotion of Science.
Author to whom correspondence should be addressed.
b͒
0021-9606/2000/113(16)/6598/10/$17.00
6598
© 2000 American Institute of Physics

J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Photodissociation of OCS
6599
FIG. 2. Schematic diagram of the experimental setup for the pump and
probe measurements using two tunable VUV laser light sources. The pho-
1
ϩ
tolysis light source excites OCS to the three lowest resonances of the 2
⌺
1
1
ϩ
FIG. 1. The PHOFEX spectrum of the entire region of the 2
band of OCS measured in our previous study ͑Ref. 7͒.
⌺ ⌺
ϩ –1
state and the probe light source probes the CO fragments thorough the
1
ϩ
A1⌸–X
⌺
transitions. For probing the S(1S) fragments, a tunable UV
light source is used in place of the VUV light source for the probe ͑see Sec.
II͒.
the second tunable VUV light source, and probe the S(1S)
fragments by tuning the UV light near the 3D°–1S transition
of the sulfur atoms to derive the vibrational distribution of
the CO fragments through the measurements of the Doppler
profiles. On the basis of the product state distributions, we
discuss the mechanism which governs the later stage of the
photodissociation of OCS after it leaves the FC region of the
2 1⌺ϩ surface.
A tunable VUV laser light was generated by the two-
photon resonant four-wave difference frequency mixing in
the Xe gas.8 For the VUV light for the photolysis of OCS,
the output beams of the two dye lasers ͑Lambda Physik
FL3002͒ simultaneously pumped by a XeCl excimer laser
͑Lambda Physik LPX 205i͒ were used. The frequency-
doubled output (␻1) of one dye laser was tuned to the two-
In spite of the long history of the pump–probe-type in-
vestigation of molecular dynamics using the laser light
sources, as far as we know, the present study reported for the
first time the pump–probe experiments in which two sets of
tunable VUV laser light sources generated by a two-photon
resonant four-wave mixing scheme are employed. It is true
that two- or three-photon LIF or resonantly enhanced multi-
photon ionization ͑REMPI͒ techniques could be used to
probe photofragments, but a nonlinearity of the process with
respect to the laser intensities as well as complex rotational-
transition selection rules for molecular fragments causing the
overlapping rotational structure tends to introduce a large
ambiguity in the product state distribution of the fragments.
Therefore, the present success of the one-photon probe fol-
lowed by one-photon excitation in the VUV region is en-
couraging in the sense that the number of the research targets
of photodissociation spectroscopy in the VUV region could
be largely increased.
photon resonant frequency of the Xe 6p 1/2 ͑80 119.474
͓
͔
0
cmϪ1͒, and then the output of the other dye laser (␻2 :
13 700–15 385 cmϪ1͒ was tuned to generate the VUV fre-
quencies (2␻1 –␻2) corresponding to the three lowest reso-
*
nant peaks ( ϭ0–2) in the PHOFEX spectrum of the
v
2 1⌺ϩ –1 1⌺ϩ band of OCS in Fig. 1, appearing at 64 745,
65 595, and 66 380 cmϪ1.7 The two light beams with the
frequencies of ␻ and ␻ were overlapped coaxially by a
1
2
dichroic mirror and focused into a stainless steel cell contain-
ing the Xe gas ͑15–30 Torr͒ by a achromatic lens (f
ϭ300 mm). The VUV laser light generated in the Xe cell
was collimated or mildly focused by a LiF lens (f
ϭ120 mm) and separated from the ␻ and ␻ light beams by
1
2
a LiF prism placed in the separator chamber connected to the
Xe cell. Then, only the generated VUV light beam was in-
troduced into the main chamber and it crossed the free jet of
the sample gas 12 mm downstream from the nozzle orifice at
an angle of 45 degrees.
In order to probe the vibrational and rotational states of
the CO fragments, another tunable VUV laser light was gen-
erated by using a dye laser ͑Lambda Physik Scanmate͒ and
an OPO laser ͑Lambda Physik Scanmate OPPO͒ simulta-
neously pumped by third harmonics of the Nd:YAG laser
with the injection seeding ͑Coherent Infinity 40-100͒. The
II. EXPERIMENT
The schematic diagram of the experimental setup is
shown in Fig. 2. A sample gas ͓OCS ͑10%͒/Ar͔ was ex-
panded into a vacuum chamber from a pulsed valve ͑10 Hz͒
with the 0.8 mm orifice diameter with the stagnation pressure
of 1.5 atm. When the pulsed valve was operated, the pressure
in the chamber was about 1.5ϫ10Ϫ4 Torr. The rotational
temperature of OCS in the expanded free jet was estimated to
be about 6 K, which was obtained for CO contained in OCS
sample ͑Matheson, 97.5% purity͒ as the trace amount of im-
purity from the measurements of the VUV–LIF spectra of
the A1⌸–X 1⌺ϩ(0,0) transitions.
same 2␻1 –␻ scheme was adopted to generate the second
2
VUV laser to probe the CO fragments as in the generation of
the photolysis VUV laser light. The bandwidth of the VUV
laser light for both the photolysis and the probe was 0.45
cmϪ1 in FWHM.
The probe VUV laser light was introduced into the
chamber so that it hits the crossing point of the molecular

6600
J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Itakura, Hishikawa, and Yamanouchi
beam and the photolysis VUV laser beam at an angle of 45
degrees with the molecular beam and at right angles with the
photolysis VUV laser beam. The direction of the polarization
of the photolysis VUV laser beam was set to be parallel with
that of the probe VUV laser beams. The fluorescence was
monitored by a solar-blind photomultiplier ͑Hamamatsu
R1259͒ placed beneath the triple crossing point through a
light-collection optics composed of a pair of LiF lenses (f
ϭ30 mm). By scanning the wavelength of the probe VUV
laser, the LIF spectra of the A1⌸–X 1⌺ϩ(0,0) ͑64 640–
64 800 cmϪ1͒ and ͑1, 1͒ ͑63 900–64 150 cmϪ1͒ bands of CO
were measured.
In order to probe the S(1S) fragments produced from the
photolysis of OCS by the VUV light, the output of the dye
laser ͑Lambda Physik Scanmate͒ pumped by third harmonics
of a Nd:YAG laser with injection seeding ͑Coherent Infinity
40-100͒ was frequency doubled in a BBO crystal, and tuned
around the frequency of the 3D°–1S transition ͑45 636
cmϪ1͒ of the S atoms. The probe UV laser beam was intro-
duced into the main chamber so that it crosses with the pho-
tolysis VUV laser beam and the free jet in the same manner
as the probe VUV laser for monitoring the CO fragments.
By monitoring the atomic fluorescence at 148 nm emit-
3
3
ted from the D° state to the P state of the S atoms, the
frequency of the probe laser was scanned to measure the
Doppler profile of the S(1S) fragments. The intracavity eta-
lon was used for the probe laser to make the bandwidth
͑FWHM͒ as narrow as 0.08 cmϪ1. In order to set the polar-
ization of the photolysis VUV laser to be parallel with the
direction of the propagation of the probe UV laser, the direc-
1
FIG. 3. The observed LIF spectra for the ͑a͒ A1⌸–X
⌺
ϩ(0,0) and ͑b͒
tion of the polarization of the ␻ light was rotated by 90
2
A1⌸–X
⌺
ϩ(1,1) bands of CO produced via the lowest vibrational reso-
1
degrees by using a pair of Fresnel rhombs. For achieving the
high purity of the linear polarization of the VUV light, the
1
ϩ
*
nance with ϭ0 of the 2
⌺
state of OCS.
v
light beam with the frequency of ␻ was transmitted though
2
a Glan-laser polarizer after it passed through the pair of
Fresnel rhombs.
formed by the monitored photolysis and probe laser intensi-
ties.
For the purification of the OCS sample to eliminate im-
purities of CO, several freeze–pump–thaw cycles were per-
formed at the liquid N2 temperature. After the purification it
was confirmed that the LIF signal of the impurity CO did not
appear with detectable intensities except the transition from
III. RESULTS AND DISCUSSION
A. Rotational state distribution of CO „X 1⌺¿;
vCOÄ0,1…
the jϽ3 rotational levels of the COϭ0 state.
v
1. Determination of rotational state distribution
The probe laser beam was introduced with a certain de-
lay after the photolysis laser radiation to reduce the contri-
bution from the scattered light of the photolysis laser and/or
the fluorescence from the impurity CO excited by the pho-
tolysis laser. The delay time was about 30 ns when monitor-
ing S(1S) and about 160 ns when monitoring CO (X 1⌺ϩ).
The intensities of the respective photolysis and probe
VUV laser beams were monitored after they passed through
the molecular beam by being reflected by a quartz plate and
guided to a solar-blind photomultiplier ͑Hamamatsu R1259͒
placed in the housing connected to the vacuum chamber. The
probe UV laser intensity was monitored by detecting the
weak reflex from a surface of a prism or a window by using
a photodiode ͑Hamamatsu S1336-5BQ͒ before the UV laser
was introduced into the vacuum chamber. The normalization
of the fluorescence intensity from the fragments was per-
After OCS was selectively excited to the quasi-bound
Ϫ1
resonances at 64 745 cmϪ1
(
ϭ0), 65 595 cm
(
*
*
ϭ1),
v
v
and 66 380 cmϪ1
(
*
ϭ2), the LIF spectra of the
v
A1⌸–X 1⌺ϩ(0,0) and ͑1,1͒ transitions of the CO fragments
were measured. The observed LIF spectra exhibiting well-
resolved rotational structure are shown in Figs. 3͑a͒ and 3͑b͒.
On the basis of the rotational assignments, the rotational-
state distributions of the CO fragment in the COϭ0 and 1
v
states were derived using the unperturbed rotational lines by
dividing the observed rotational transition intensities by the
¨
corresponding Honl–London factors. The derived rotational
state distributions of the COϭ0 and 1 states of the CO frag-
v
ments produced through the three lowest resonances of OCS
are plotted in Fig. 4. It should be noted in Figs. 4͑a͒–4͑f͒ that
the populations derived from the Q branch transitions were
always smaller than those derived from the P and R branch

J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Photodissociation of OCS
6601
OCS is prepared under the ultracold conditions at the very
low rotational temperature. The magnitude of the total angu-
lar momentum, J, is considered to be very small and could
be assumed to be zero, i.e., Jϭ0, for simplification. When
Jϭ0, the recoil vector, v, is constrained to be on the molecu-
lar plane defined by the three atoms and fixed in space during
the dissociation process and both j and L(ϭϪj) become
perpendicular to the plane in the high jϭ͉j͉ limit. Since the
2 1⌺ϩ –1 1⌺ϩ transition is a parallel transition, the transition
moment, ␮OCS , is located in the molecular plane. Therefore,
the correlations among the aforementioned vectors are rep-
resented as vЌj and ␮ Ќj.
OCS
In the high-j limit, the transition moment of CO, ␮
,
CO
for the A1⌸–X 1⌺ϩ transition is considered to be parallel to
j for the Q branch transition and perpendicular to j for the P
and R branches. In our experimental condition, the direction
of the polarization of the photolysis laser light was parallel to
the direction of the polarization of the probe light for the CO
fragments. Therefore, the LIF intensities in the P and R
branch transitions of the CO fragments should be enhanced
compared with those in the Q branch transitions due to the
anisotropic angular momentum distribution of CO.
By taking account of the ␮OCS–v–j vector correlations
and the detection geometrical factor,9,11 the Doppler line pro-
files for the respective P, Q, and R branch transitions were
calculated in a semi-classical manner described by Dixon.9
In the calculation, the recoil anisotropy parameter was as-
sumed to be ␤ϭ1.8 for all the rotational states of the CO
fragments, which was previously derived for the photodisso-
ciation of OCS at 157 nm.13 After convoluting the calculated
Doppler profiles by a Gaussian function with the probe laser
bandwidth (FWHMϳ0.45 cmϪ1), the rotational spectrum
was synthesized, which can be compared with the observed
rotational spectrum. Then, by dividing the observed peak
intensities in the LIF spectrum by the simulated rotational
transition intensities of the corresponding peaks, the cor-
rected rotational state distributions of the CO fragments were
obtained as shown in Fig. 5. The extent of the discrepancies
in Fig. 4 between the rotational distributions obtained from
the P and R branch transitions and those obtained from the Q
branch transitions decreases considerably in Fig. 5, espe-
FIG. 4. The observed rotational state distributions of the CO fragments in
the vibrational states of COϭ0 and 1 via the three lowest vibrational reso-
v
¨
*
nances of
ϭ0–2, derived by taking account only of the Honl–London
v
factor from the peak intensities of the P͑᭡͒, Q͑᭺͒, and R͑᭿͒ branches in the
observed LIF spectra. The length of the bar for the data points represents the
experimental uncertainty ͑␴͒.
transitions. As explained later, these discrepancies are as-
cribed to the alignment effect of the CO fragments induced
by the polarized photolysis laser light, i.e., the angular mo-
mentum vector of the CO fragment is formed preferentially
in the direction perpendicular to the polarization vector of
the photolysis light.9–12
cially in the high-j region of the COϭ1 channels. However,
v
even after the correction, there still remain some discrepan-
cies, and the populations derived from the Q branch transi-
tions tend to become larger than those derived from the P
and R branch transitions. This tendency is most clearly seen
The lifetimes of OCS in the lowest three quasi-bound
vibrational resonances of 2 1⌺ϩ were determined to be
*
*
*
v
133 fs ( ϭ0), 44 fs ( ϭ1), and 27 fs ( ϭ2) in our
v
v
previous study.7 These three values are all much shorter than
a rotational period of OCS, and therefore, the CO photofrag-
ments could not recoil isotropically with respect to the po-
larization direction of the photolysis laser light within such
an ultra-short time duration. In such an ultrafast dissociation
process, a phenomenon originating from the correlation
among ͑i͒ the polarization vector of the photolysis light, ͑ii͒
the recoil vector, and ͑iii͒ the angular momentum of the frag-
ment can be observed.
in the low-j region of the COϭ0 channel. Although the
v
value of ␤ was changed in the simulation to reduce the dis-
crepancies in the low-j region, these were not improved in
any value of ␤.
The discrepancies between the populations obtained
from the Q branch and those from the P and R branches
could be attributed to ͑i͒ the assumption of Jϭ0 and ͑ii͒ the
semi-classical approximation. Even under the jet-cooled con-
dition (Trotϭ6 K), the population of OCS in the rotational
ground state with Jϭ0 is only ϳ5%. Consequently, the vec-
From the conservation of the angular momentum, the
total angular momentum of OCS, J, before the dissociation is
equal to the sum of the angular momentum of the CO frag-
ment, j, and the orbital angular momentum, L, of the recoil-
ing fragments of S and CO, i.e., JϭjϩL. In our experiment,
tor correlations of vЌj and ␮ Ќj do not always hold.12
OCS
When the angular momentum of the CO fragment, j, is much
larger than J, the semi-classical approximation becomes ap-

6602
J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Itakura, Hishikawa, and Yamanouchi
TABLE I. The rotational temperatures ͑in K͒ for the CO fragments in the
COϭ0 and 1 states produced through the three lowest vibrational reso-
v
nances of ϭ0–2 of OCS in the 2
⌺
state.a
1
ϩ
*
v
*
COϭ0
v
COϭ1
v
v
0
1
2
95 ͑10͒
100 ͑30͒
90 ͑15͒
2210 ͑280͒
940 ͑120͒
810 ͑65͒
aThe numbers in the parentheses represent a standard deviation ͑␴͒ derived
from the least-squares fit to the Boltzmann-type rotational distribution.
tional distributions for the COϭ0 and 1 channels for the
v
*
ϭ0–2 resonances obtained from the P and R branch tran-
v
sitions were fitted by the least-squares method to derive the
rotational temperatures of the CO fragments. The determined
rotational temperatures are listed in Table I.
*
For the photodissociation via
(ϭ0–2), the rotational
v
temperatures of the CO fragments in COϭ0 are all very
v
low, i.e., Trotϳ100 K, while those in COϭ1 are much
v
higher than in COϭ0, i.e., Trotϭ2210, 940, and 810 K for
v
*
ϭ0, 1, and 2 respectively, and exhibit a tendency to be-
v
*
come lower as
increases.
v
2. Comparison with model distributions
a. Prior distribution. We will compare the observed ro-
tational distributions with prior distributions, in which the
CO fragments are populated with probabilities proportional
to a square root of the translational energy in all the ener-
getically allowed ro-vibrational states.14,15 If the total angular
momentum J of a parent triatomic molecule is approximated
to be zero in its dissociation process, the angular momentum
of the diatomic fragment, j, is perpendicular to the direction
of the recoil, and the projection of j onto the recoil direction,
mv , becomes zero. Therefore, when calculating a prior dis-
FIG. 5. The corrected rotational state distributions of the CO fragments
obtained from Fig. 4 by taking account of the vector correlations among the
photolysis laser polarization, the fragment recoil, and the angular momen-
tum of the CO fragments.
tribution function, P ( CO , j), of the CO fragment in a
v
0
single vibrational and rotational state ( CO , j), the (2jϩ1)
v
propriate and the deviations from the vector correlations tend
to be small. However, in the low-j region, it would not be
appropriate any more, and this is the reason why the discrep-
ancies between the populations obtained from the Q branch
and those from the P and R branches are larger in the low-j
manifold as shown in Fig. 5 even after the correction con-
sidering the vector correlations.
degeneracy of the rotational state should not be taken into
account,16 and it is represented by a square root of the trans-
lational energy as
1/2
P0
, j͒ϰ EavailϪEvϪEr͒
͑2͒
͑vCO
͑
with
On the other hand, as long as we concentrate on the
distributions derived from either the Q branch or the P and R
branches, the effect originating from the correction consider-
ing the vector correlations was found to be negligibly small.
Because the number of rotational states whose populations
were determined from the P and R branches were much
larger than those from Q branch lines, the P and R branch
transitions were adopted to derive the rotational distribu-
tions, which were obtained simply by dividing the rotational
E
availϭEh␯ϪD0ϪES ,
͑3͒
where Eh␯ is the energy of the photolysis light, D0 is the
dissociation energy of OCS in the electronic ground state, ES
is the electronic energy of the S state of the S atom mea-
sured from the P2 state, and Ev and Er are the vibrational
and rotational energies of the CO fragment, respectively.
From the rotational state distribution of the CO frag-
1
3
ments in their specific vibrational state, P( CO , j), derived
v
experimentally in Sec. III A 1, and the prior distribution,
¨
transition intensities by the Honl–London factors.
P ( CO , j), the rotational surprisal, I(gR), is obtained14–17
v
0
By assuming that the rotational distribution of the CO
fragment, C(j), as that expressed by the Boltzmann distri-
bution,
as
P
CO , j͒
͑v
I g ͒ϭϪln
͑
͑4͒
R
C j͒ϭC 2jϩ1͒ exp ϪERot /kbTrot͒,
͑1͒
P0 CO , j͒
͑
͑
͑
͑v
0
with a proportional constant C0 independent of j, the rota-
as a function of

J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Photodissociation of OCS
6603
*
the P and Q branch data for
ϭ0 and 1 exhibit a shallow
v
minimum at gRϭ0.06–0.08 and gRϭ0.02–0.05, respec-
*
tively. For ϭ2, it would be hard to describe the tendency
v
due to the small data number in the plot, but the P branch
data above gRϭ0.015 exhibit a clear positive slope. It is
*
uncertain whether the
ϭ2 data have a minimum, but it
v
can be said that the transition from the flat region to the
positive slope region occurs at gRϭ0.01–0.02. The complex
behavior of the surprisal plots described above for the rota-
tional distributions of the CO fragments in the COϭ1 state,
v
suggests that a simple dynamical constraint could not de-
scribe the rotational distributions. It is probable that two dif-
ferent kinds of dynamics govern respectively the production
of the low-j and high-j CO fragments.
b. Rotational Franck–Condon model. When the PES of
the excited state of OCS around the FC region is flat along
the bending angle and no torque is imposed on the CO frag-
ments to drive their rotational motion, the rotational state
distribution is determined by the projection of the bending
wave function of OCS in the electronic ground state onto the
rotational wave function of the CO fragments. Within the
framework of this rotational FC model, Freed and Morse18
showed that the rotational distribution of diatomic fragments
can be represented by a Boltzmann distribution when photo-
dissociation of a linear triatomic molecule occurs from its
Jϭ0 and bendϭ0 state. By adopting the rotational FC
v
model, Strauss et al.13 calculated the rotational temperature,
FIG. 6. The rotational surprisal plots, I(gR)s, for the CO fragments in the
T
rot , of Trotϭ380 K for the CO fragments produced through
OCS (2 1⌺ϩ). However, the PES of the 2 1⌺ϩ state of OCS
could have a certain slope along the bending direction
around the FC region which induces the bending motion of
OCS and eventually imposes the torque for the rotation on
CO. Whether the rotation of the CO fragment is promoted or
suppressed depends on the anisotropy of the 2 1⌺ϩPES
along the bending angle in the course of the dissociation.
Therefore, the deviation from the Boltzmann distribution
with Trotϭ380 K, derived from the rotational FC model,
could be ascribed to the anisotropy of the PES. Since the CO
vibrational states of COϭ0 and 1 produced via the quasi-bound vibrational
v
*
resonances of ϭ0, 1, and 2. The data derived from the intensities of the
v
observed P͑᭡͒, Q͑᭺͒, and R͑᭿͒ branches in the LIF spectra are plotted with
different marks. The straight lines drawn in ͑a͒–͑c͒ represent the best-fit
linear lines for the surprisal plots for
ϭ0. For COϭ1, the three plots do
vCO
v
not exhibit simple linear behavior and, therefore, no straight line is drawn.
Er
Er /Eavail
fR
gRϭ
ϭ
ϭ
.
͑5͒
E
availϪEv 1Ϫ E /E
͒
avail
1Ϫfv
͑
v
The resulting plots of the rotational surprisal are shown in
Fig. 6 for the CO fragments in the two lowest vibrational
fragments with COϭ0 have a rotational temperature of
around 100 K for all the three low-lying resonances of
ϭ0–2, the CO fragments with COϭ0 are considered to be
formed through the anisotropic PES, which tends to suppress
the rotation of the fragments. On the other hand, the CO
fragments with COϭ1 exhibit much higher rotational tem-
peratures than Trotϭ380 K; i.e., 2210, 940, and 810 K for
v
v
*
v
states ( COϭ0,1) when the three lowest vibrational reso-
v
v
*
nances ( ϭ0–2) of the parents OCS molecules are ex-
v
cited. Then, the surprisal plots were fitted by a linear func-
tion to derive a slope. In the case of the CO fragments with
v
COϭ0, the slopes obtained from the P branch transitions are
v
*
119͑32͒, 149͑42͒, and 147͑31͒ for
ϭ0, 1, and 2, respec-
v
*
ϭ0, 1, and 2, respectively. Therefore, the CO fragments
tively, while those obtained from the Q-branch are 123͑23͒,
100͑7͒, and 145͑24͒. Even though the slopes are associated
with relatively large uncertainties, it can be said that the plots
are well described by a linear function and that they exhibit
consistently a slope of 100–150. This suggests that there
exists a common mechanism which governs the rotational
with COϭ1 are considered to be formed through the aniso-
v
tropic PES, which tends to promote the rotation of the frag-
ments though the extent of the rotational excitation is sup-
pressed for the higher resonance state of OCS.
distribution in the COϭ0 for the three lowest resonances.15
3. Effect of resonances in the FC region
v
In the case of the surprisal plots for COϭ1, their behav-
Previously, Strauss et al.13 studied the rotational state
distribution of the CO fragments produced through the pho-
tolysis of OCS at the fixed wavelength of 157 nm. The rota-
v
ior is not as simple as that for COϭ0. In the range below
v
g ϭ0.015, where the data for both COϭ0 and 1 are avail-
v
R
*
able, the P branch data for ϭ0 and 2 of COϭ1 exhibit an
tional distributions for COϭ0–3 were found to be well re-
v
v
v
almost flat distribution with respect to gR , while those for
produced respectively by the Boltzmann distributions with
*
ϭ1 exhibit a negative slope. When it comes to the global
the rotational temperatures of Trotϭ1350, 1300, 980, and 770
v
behavior of the plot in the entire range, 0рgRр0.15, both
K. This decreasing tendency of Trot for a larger CO shows a
v

6604
J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Itakura, Hishikawa, and Yamanouchi
marked contrast with the present results obtained at the three
lowest vibrational resonances for which the substantially
higher rotational temperatures ͑1000–2000 K͒ are achieved
for
ϭ1 than for COϭ0(ϳ100 K).
vCO
v
As was reported in our previous study,7 a weak con-
tinuum is located at 157 nm under the jet-cooled conditions.
*
At the peaks of the vibrational resonances ( ϭ0–2) there
v
is a large contribution from the resonance component repre-
senting the nuclear motion along the direction perpendicular
to the dissociation coordinate, while at 157 nm there is
mostly the off-resonance component representing the nuclear
motion along the dissociative coordinate. Therefore, the ob-
served difference in the
dependence of the rotational
vCO
temperature could be attributed to the difference in the char-
acter of the nuclear motion near the FC region on the upper
state PES.
As shown in Fig. 1, the vibrational resonances in the
PHOFEX spectrum of the 2 1⌺ϩ –1 1⌺ϩ band exhibit char-
*
acteristic peak profiles; the ϭ0 resonance shows a largely
v
asymmetrical profile shading towards the lower energy side
with a narrow peak width, while the peak profile becomes
*
*
more symmetrical and broader as
increases to the
v
v
ϭ1 and 2 resonances. It would be natural to expect that such
a character of the resonances influences the product state
distributions, because the extent of the asymmetry as well as
the peak width reflects the extent of the mixing of the con-
tinuum and bound characters.19 However, it seems that the
memory of the character of the resonance state is mostly lost
in the rotational state distribution for the COϭ0 channel,
v
suggesting the existence of a strong exit-channel interaction
between the FC region and the asymptotic region. The loss
of the memory of the resonance states is also seen in the
vibrational state distributions of CO described in the next
section. On the other hand, the extent of the rotational exci-
*
in-
tation for the
ϭ1 channel is more suppressed as
vCO
v
creases from 0 to 2 ͑see Table I and Fig. 6͒. It is possible that
the remnants of the character of the resonances appear in the
form of the difference in the rotational excitation for the
FIG. 7. The observed ͑open circles͒ and best-fit ͑solid lines͒ Doppler pro-
1
files of the S(1S) fragments for the photodissociation of OCS (2
⌺
ϩ) via
*
*
*
v
the quasi-bound states of ͑a͒ ϭ0, ͑b͒ ϭ1, and ͑c͒ ϭ2. In the trial-
v
v
COϭ1 channel.
v
and-error simulation, the vibrational distributions of the CO fragments were
assumed to be represented by the Boltzmann-type distribution and the same
optimized vibrational temperatures of Tvibϭ7000 K were derived for all the
three profiles ͑see text for details͒.
B. Vibrational state distribution of CO „X 1⌺¿…
profile, the UV probe laser light to detect the S(1S) frag-
ments was introduced along the direction parallel with the
electric vector of the photolysis laser light so that the two
laser beams cross at right angles with each other.
The observed Doppler profiles obtained by exciting the
central position of the peak for the three lowest resonant
1. Vibrational state distribution of CO from Doppler
1
profiles of S„ S…
In order to derive the vibrational state distribution of the
CO fragments, we measured the Doppler profiles of the
1
S(1S) fragments. Since a S state has no electronic angular
momentum, the Doppler profile reflects directly the recoil
momentum of the S(1S) fragments,9,10,20 from which the ki-
netic energy distribution of the S(1S) fragments is obtained.
Then, on the basis of the conservation of the energy and the
momentum, the internal energy distribution of the CO frag-
ments including their vibrational state distribution can be
estimated.
*
*
v
states, i.e., 64 745( ϭ0), 65 595( ϭ1), and 66 380
v
cmϪ1
(
*
ϭ2), are shown in Figs. 7͑a͒–7͑c͒. All the Dop-
v
pler profiles exhibit a clear dip in the center, indicating that
the released energies in the dissociation process are effi-
ciently transferred into the translational motion rather than
into the rotational and vibrational degrees of freedom of the
CO fragments.
From the rotational analysis of the LIF spectra of the
A 1⌸–X 1⌺ϩ(0,0) and ͑1, 1͒ bands of the CO fragments in
Sec. III A 1, the rotational state distributions of the CO frag-
In the case of the 2 1⌺ϩ –1 1⌺ϩ transition,21,22 the elec-
tronic transition moment is parallel to the molecular axis,
and the fragments are expected to eject preferentially in the
direction along the laser polarization. Therefore, in order to
observe the Doppler effect in a form of a clearer Doppler
ments in the vibrational states of COϭ0 and 1 were found to
v

J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Photodissociation of OCS
6605
FIG. 9. The normalized vibrational state distributions of the CO fragments
1
produced after the photodissociation of OCS (2
distribution of 7000 K which reproduces all the observed Doppler profiles of
⌺
ϩ); ͑a͒ the Boltzmann
the S(1S) fragments when the ϭ0–2 resonances of OCS are excited, ͑b͒
*
v
*
the prior distribution calculated for the ϭ0 resonance, and ͑c͒ the vibra-
v
tional distribution for COϭ0–3 obtained by Strauss et al. ͑Ref. 13͒ for the
v
photodissociation at 157 nm. For the normalization for ͑c͒, the amount of
the population in each of
ϭ4–8 is assumed to be 5% of COϭ0, which
vCO
v
was estimated in Ref. 13 as an upper limit of the population for these five
vibrational states.
seen in Figs. 8͑a͒ and 8͑b͒ that the contribution from the
highly excited CO must be relatively small to reproduce the
*
central dip in all the observed Doppler profiles for
v
ϭ0–2.
In the course of the simulation of the Doppler profiles, it
was found that the Boltzmann-type distribution in which the
vibrational population exponentially decreases as the vibra-
tional energy increases was found to give the best fit. After a
trial-and-error-type simulation using the vibrational tempera-
ture, Tvib , as one variable parameter, the optimized vibra-
tional temperature of Tvibϭ7000Ϯ2000 K was obtained for
FIG. 8. ͑a͒ The observed Doppler profile of the S(1S) fragments when OCS
*
was excited to the lowest quasi-bound resonances with
ϭ0 ͑solid line͒.
v
The calculated Doppler profiles are shown for comparison for the lowest
vibrational state ( ϭ0) and the highest vibrational state ( COϭ8) ͑dotted
vCO
v
lines͒ when the CO fragments are assumed to be populated only in the single
vibrational state. ͑b͒ The calculated Doppler profiles are shown for all the
vibrational states, COϭ0–8, which are energetically accessible.
v
*
all the three resonances of ϭ0–2. As shown in Fig. 7, the
v
best-fit Doppler profiles reproduced well the observed pro-
files. This result suggests that there exists a common mecha-
nism for all the three resonances which governs the vibra-
tional population of CO by affecting in a later stage after the
FC region regardless of the different excitation energies.
be approximated by the Boltzmann distributions for the dis-
sociation through the three low-lying resonance states with
*
ϭ0–2. Therefore, in the simulation of the Doppler pro-
v
files of the S(1S) atomic fragments, the rotational distribu-
tions for the energetically accessible vibrational states of CO
were assumed to be described by the Boltzmann distribu-
tions. The rotational temperatures for the vibrational states
2. Comparison with prior distributions
A prior distribution of vibrational states is obtained by
the summation of P ( CO , j) over all the rotational states
which are energetically accessible as
v
except for COϭ0 and 1 were assumed to be the same as the
v
0
corresponding values for COϭ1; i.e., Trotϭ2210, 940, and
v
810 K were adopted for the fragment CO ( COϾ1) produced
v
*
through the three lowest resonances with
ϭ0, 1, and 2,
v
P0
͒ϭ
P
CO , j͒.
͑v
͑6͒
͑vCO
͚
0
j
respectively. The recoil anisotropy parameter ␤ was assumed
again to be ␤ϭ1.8.
In Fig. 9, the best-fit Boltzmann vibrational state distribution
of Tvϭ7000 K, Pobs CO), to describe the observed Doppler
profile is compared with the prior distribution calculated for
The Doppler profile of S(1S), obtained when OCS is
(
v
*
excited to the lowest vibrational resonance of
shown in Fig. 8͑a͒, and the Doppler profiles calculated as-
suming that the CO fragments are populated in a single vi-
brational state in the range between COϭ0 and 8 are shown
in Fig. 8͑b͒. For comparison with the observed Doppler pro-
file, the calculated Doppler profiles for COϭ0 and 8 are also
drawn in Fig. 8͑a͒ in dotted lines. When OCS is excited to
ϭ0, is
v
*
the ϭ0 resonant state of OCS. It can be seen in this figure
v
that the CO fragments tend to be populated in the lower
vibrational states than in the prior distribution. The discrep-
v
ancy between the observed distribution, Pobs
( CO), and the
v
v
prior distribution, P ( CO), is described by the vibrational
v
0
surprisal, I(fv),14,17
*
ϭ0 and 1, the maximum vibrational quantum number of
v
P
͒
͑vCO
CO, which can be produced energetically, is COϭ8, and
v
I f ͒ϭϪln
,
͑7͒
͑
v
P0
͒
͑vCO
*
when
ϭ2, the maximum becomes COϭ9. It is clearly
v
v

6606
J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Itakura, Hishikawa, and Yamanouchi
formed in the cold region are considered to be sampled, the
vibrational state distribution reported in Ref. 13 could be
regarded as that obtained when the off-resonant continuum is
excited. It is interesting to note that the photodissociation via
*
the three resonances, ϭ0–2, afforded a lower vibrational
v
temperature than the photodissociation at 157 nm. It can be
said that the excitation of the vibration of the CO fragments
is suppressed when the dissociation proceeds through the
resonance states of the quasi-bound vibration in the FC re-
gion, while the energy partitioning into the fragment vibra-
tion becomes efficient when the off-resonant continuum is
excited.
The observation that the similar vibrational distributions
*
are obtained for all the three resonance states,
ϭ0–2,
v
indicates that the quasi-bound vibration in the FC region
would not affect so much the vibrational distribution of the
CO fragments. This means that the product state distribution
is determined in the region between the FC region and the
asymptotic region. In other words, the memory of the vibra-
tion in the FC region tends to be lost when the dissociation
proceeds.
In the cases of the photodissociation of FNO (S1)23,24
and ClNO (T1 ,S1),25–27 their PHOFEX spectra exhibited
distinct vibrational peaks and they were interpreted as the
vibrational Feshbach resonances. Therefore, the vibrational
dynamics of FNO and ClNO in their FC region is considered
to have a certain similarity to OCS (2 1⌺ϩ). However, con-
trary to OCS, the vibrational distributions of the diatomic
photofragments produced from FNO and ClNO were inter-
preted by an adiabatic picture, in which most of the NO
fragments produced via the vibrational Feshbach resonance
FIG. 10. The vibrational surprisal plots, I(fv)s, for the CO fragments ob-
*
*
*
ϭ1, and ͑c͒
v
tained via the quasi-bound resonances; ͑a͒
ϭ0, ͑b͒
v
v
ϭ2. The slopes of the fitted lines to the surprisal plots in the range between
*
ϭ0, 1, and 2,
v
COϭ0 and 6 are 1.90͑13͒, 2.14͑12͒, and 2.36͑13͒ for
respectively.
v
*
with the quantum number,
, are populated in the vibra-
v
as a function of fvϭEv /Eavail . The surprisal plots of the CO
fragments obtained for the three lowest resonances of OCS
*
v
tional state with the quantum number,
, equal to
or
vNO
*
Ϫ1. This type of an adiabatic distributions suggests that
v
*
with
ϭ0–2 are presented in Fig. 10. All three surprisal
v
the vibrational motion of the NO moiety of the parent mol-
ecules in the FC region is very close to the vibrational mo-
tion of the NO fragments and keeps its nature of local vibra-
tion in the course of the dissociation process.
plots exhibit a linear slope in the range of the vibrational
states of CO with COϭ0–6, to which most of the fragments
v
are populated. By the fit to a straight line, the slopes were
*
determined to be 1.90͑13͒, 2.14͑12͒, and 2.36͑13͒ for
v
As far as the vibrational distributions of the diatomic
fragments are concerned, the photodissociation of OCS
(2 1⌺ϩ) are closer to the symmetric triatomic molecules of
ϭ0, 1, and 2, respectively. This linear behavior commonly
identified for the dissociation through the three low-lying
resonances with the close values of the best-fit slopes indi-
cates the existence of a common mechanism which governs
the vibrational energy partitioning process.15
28–32
1
˜
H2O (A, B1)
and O3 ͑Chappuis Band33–35 and Hartley
Band36–39͒ rather than the X–NO-type molecules even
though the broad absorption features of H2O and O3 exhibit
no vibrational progression. When these symmetric triatomic
molecules are excited to the dissociative PES of the elec-
tronically excited state, they start to move along the symmet-
ric stretch coordinate from the FC point towards the saddle
point. Though the potential along the symmetric stretch co-
ordinate is bound, this motion is unstable and is deformed
into the mixture of the translational motion along the anti-
symmetric stretch coordinate and the vibration of the di-
atomic fragment, which is expected to cause the breakdown
of the vibrational adiabaticity.
3. Effect of resonances in the FC region
As shown in Fig. 9, the calculated vibrational distribu-
tion of CO at 7000 K, which reproduces well the observed
Doppler profiles obtained via the vibrational resonances of
*
ϭ0–2, have the larger populations in the low vibrational
v
states than the vibrational distributions obtained from the
photodissociation at 157 nm by Strauss et al.13 Because the
*
available energy at ϭ0–2 is larger than that obtained by
v
the 157 nm photodissociation, this difference is contrary to
the statistical model in which the population in the higher
vibrational states of the fragments increases as the available
energy becomes larger.
In the case of OCS, the vibrational motion near the FC
region is considered to be close to the in-phase stretching in
both the O–C and C–S bonds as was demonstrated by the
wavepacket calculation in our previous study.7 Therefore, in
the course of the dissociative process, the vibrational motion
Since the photolysis at 157 nm by Strauss et al.13 was
performed under jet-cooled conditions and the CO fragments

J. Chem. Phys., Vol. 113, No. 16, 22 October 2000
Photodissociation of OCS
6607
in the FC region would be transferred into both the transla-
tional motion and the vibration of the CO fragments, and
during this process the vibrational adiabaticity would be lost,
resulting in the similar vibrational state distribution to O3 and
H2O.
been supported by the Grant-in-Aid for Scientific Research
͑No. 08640636͒ and that for Priority Area ͑No. 07240106͒
from the Ministry of Education, Science, Sports and Culture
and by CREST ͑Core Research for Evolutionary Science and
Technology͒ fund from Japan Science and Technology Cor-
poration. R.I. acknowledges the fellowship from Japan Soci-
ety for the Promotion of Science.
IV. CONCLUSION
We investigated the photodissociation of OCS via the
2 1⌺ϩ state by measuring the rotational and vibrational state
distributions of the CO fragments produced through the three
lowest vibrational resonances in the 2 1⌺ϩ –1 1⌺ϩ band.
Since the dissociation, OCS (2 1⌺ϩ)→CO(X 1⌺ϩ)
ϩS(1S), proceeds on a single PES of the 2 1⌺ϩ state, this
band system affords us an ideal opportunity to study the pure
nuclear dynamics in the course of the dissociation process.
1 R. Schinke, Photodissociation Dynamics ͑Cambridge University Press,
New York, 1993͒.
2 E. J. Heller, Acc. Chem. Res. 14, 368 ͑1981͒.
3 C. D. Pibel, K. Ohde, and K. Yamanouchi, J. Chem. Phys. 101, 836
͑1994͒.
4 K. Yamanouchi, K. Ohde, A. Hishikawa, and C. D. Pibel, Bull. Chem.
Soc. Jpn. 68, 2459 ͑1995͒.
5 K. Ohde, A. Hishikawa, and K. Yamanouchi, J. Electron Spectrosc. Relat.
Phenom. 79, 433 ͑1996͒.
6 K. Yamanouchi, K. Ohde, and A. Hishikawa, Adv. Chem. Phys. 101, 789
͑1997͒.
͑1͒ The rotational distributions of the CO fragments in the
7 A. Hishikawa, K. Ohde, R. Itakura, S. Liu, K. Yamanouchi, and K. Ya-
mashita, J. Phys. Chem. A 101, 694 ͑1997͒.
COϭ0 and 1 vibrational states were derived for the
v
*
three lowest resonance peaks,
ϭ0–2, of OCS by
v
8 K. Yamanouchi and S. Tsuchiya, J. Phys. B 28, 133 ͑1995͒.
9 R. N. Dixon, J. Chem. Phys. 85, 1866 ͑1986͒.
measuring the rotational structures of the LIF spectra of
the one-photon CO A 1⌸–X 1⌺ϩ(0,0) and ͑1, 1͒ transi-
10 P. L. Houston, J. Phys. Chem. 91, 5388 ͑1987͒.
11 C. H. Green and R. N. Zare, J. Chem. Phys. 78, 6741 ͑1983͒.
12 G. E. Hall, N. Sivakumar, D. Chawla, P. L. Houston, and I. Burak, J.
Chem. Phys. 88, 3682 ͑1988͒.
tions. The rotational state distributions in COϭ0 for
v
*
ϭ0–2 were all represented approximately by the
v
13 C. E. Strauss, G. C. McBane, P. L. Houston, I. Burak, and J. W. Hepburn,
J. Chem. Phys. 90, 5364 ͑1989͒.
Boltzmann distribution of around 100 K. On the other
hand, the CO fragments in COϭ1 were much more ex-
v
14 R. D. Levine and R. B. Bernstein, Acc. Chem. Res. 7, 394 ͑1974͒.
15 J. I. Steinfeld, J. S. Francisco, and W. L. Hase, Chemical Kinetics and
Dynamics ͑Prentice–Hall, Englewood Cliffs, NJ, 1989͒.
16 J. T. Muckerman, J. Phys. Chem. 93, 179 ͑1989͒.
17 E. Zamir, R. D. Levine, and R. B. Bernstein, Chem. Phys. Lett. 107, 217
͑1984͒.
cited rotationally than CO ( COϭ0) and the distribu-
v
tions were approximated by the Boltzmann distributions
with Trotϭ2210Ϯ280, 940Ϯ120, and 810Ϯ65 K for
*
ϭ0, 1, and 2, respectively. From the comparison with
v
the rotational temperatures of the CO fragments obtained
by the photolysis at 157 nm and those by the rotational
FC model, the characteristic rotational temperatures de-
rived here were considered to reflect the character of the
vibrational resonances in the FC region. The difference
18 M. D. Morse and K. F. Freed, J. Chem. Phys. 74, 4395 ͑1981͒.
19 U. Fano, Phys. Rev. 124, 1866 ͑1961͒.
20 R. N. Zare, Mol. Photochem. 4, 1 ͑1972͒.
21 J. W. Rabalais, J. M. McDonald, V. Scherr, and S. P. McGlynn, Chem.
Rev. 71, 73 ͑1971͒.
22 M. I. McCarthy and V. Vaida, J. Phys. Chem. 90, 6759 ͑1986͒.
23 S. A. Reid, J. T. Brandon, and H. Reisler, J. Phys. Chem. 97, 540 ͑1993͒.
24 J. T. Brandon, S. A. Reid, D. C. Robie, and H. Reisler, J. Chem. Phys. 97,
5246 ͑1992͒.
in the rotational distribution of the
ϭ0 and COϭ1
vCO
v
channels was also identified as the difference in their
surprisal plots.
25 C. X. W. Qian, A. Ogai, L. Iwata, and H. Reisler, J. Chem. Phys. 92, 4296
͑1990͒.
͑2͒ From the measurements of the Doppler profiles of the
S(1S) atomic fragments, the vibrational distributions of
the CO fragments were determined and they were found
to be well approximated by the Boltzmann distribution
with the temperature of 7000Ϯ2000 K. The common vi-
brational distribution obtained for all the three lowest
resonance states with similar shapes in their surprisal
plots indicates that the dissociation channels which form
26 A. Ogai, C. X. W. Qian, and H. Reisler, J. Chem. Phys. 93, 1107 ͑1990͒.
27 C. X. W. Qian, A. Ogai, J. T. Brandon, Y. Y. Bai, and H. Reisler, J. Phys.
Chem. 95, 6763 ͑1991͒.
28 P. Anderson, G. S. Ondrey, B. Titze, and E. W. Rothe, J. Chem. Phys. 80,
2548 ͑1984͒.
29 V. Engle, V. Staemmler, R. L. Vander Wal, F. F. Crim, R. J. Sension, B.
Hudson, P. Andersen, S. Hennig, K. Weide, and R. Schinke, J. Phys.
Chem. 96, 3201 ͑1992͒.
30 D. W. Hwang, X. Yang, and X. Yang, J. Chem. Phys. 110, 4119 ͑1999͒.
31 V. Engle, R. Scihnke, and V. Staemmler, J. Chem. Phys. 88, 129 ͑1988͒.
32 M. L. Doublet, G. J. Kroes, E. J. Baerends, and A. Rosa, J. Chem. Phys.
103, 2538 ͑1995͒.
the CO fragments in the different CO values are coupled
v
strongly with each other after OCS leaves the FC region,
leading to the breakdown of the vibrational adiabaticity.
33 D. S. Moore, D. S. Bomse, and J. J. Valentini, J. Chem. Phys. 79, 1745
͑1983͒.
ACKNOWLEDGMENTS
34 H. B. Levene, J.-C. Nieh, and J. J. Valentini, J. Chem. Phys. 87, 2583
͑1987͒.
The authors are grateful to Professor Geerd H. F.
Diercksen for his valuable discussion. The authors also thank
Takashi Kawauchi, Nobuhiro Hirano, and Tetsuya Kimura
͑Marubun Co.͒ for their technical support for the Infinity-
based dye-laser and OPO-laser system and the LPX-based
dye-laser system which were used to generate the two sets of
tunable vacuum UV laser light beams. The present study has
35
¨
B. Woywod, M. Stengle, W. Domcke, H. Flothmann, and R. Schinke, J.
Chem. Phys. 87, 2583 ͑1987͒.
36 J. J. Valentini, D. P. Gerrity, D. L. Phillips, J.-C. Nieh, and K. D. Tabor,
J. Chem. Phys. 86, 6745 ͑1987͒.
C. Leforestier, F. LeQuere, K. Yamashita, and K. Morokuma, J. Chem.
37
´ ´
Phys. 101, 3806 ͑1994͒.
38 B. R. Johnson and J. L. Kinsey, Phys. Rev. Lett. 62, 1607 ͑1989͒.
39 B. R. Johnson and J. L. Kinsey, J. Chem. Phys. 91, 7638 ͑1989͒.