Photodissociation of vibrationally excited SH and SD radicals at 288 and 291 nm: the S(1D2) channel.

Abstract

Ultraviolet photodissociation of SH (X 2Pi, upsilon"=2-7) and SD (X 2Pi, upsilon"=3-7) has been studied at 288 and 291 nm, using the velocity map imaging technique to probe the angular and speed distributions of the S(1D2) products. Photodissociation cross sections for the A 2Sigma+<--X 2Pi(upsilon") and 2Delta<--X 2Pi(upsilon") transitions have been obtained by ab initio calculations at the CASSCF-MRSDCI/aug-cc-pV5Z level of theory. Both the experimental and theoretical results show that SH/SD photodissociation from X 2Pi (upsilon"<or=7) proceeds via the repulsive wall of the A 2Sigma+ state. The angular distributions of S(1D2) indicate that the dissociation approaches the sudden recoil limit of the A 2Sigma+ state, yielding strongly polarized fragments. The S(1D2) atoms are predominantly produced with total electronic angular momentum perpendicular to the recoil axis.

Full text

Page 1

Photodissociation of vibrationally excited SH and SD radicals
at 288 and 291 nm: The S„1D2… channel
Liesbeth M. C. Janssen, Mark P. J. van der Loo, and Gerrit C. Groenenboom
Theoretical Chemistry, Institute for Molecules and Materials (IMM), Radboud University Nijmegen,
Toernooiveld 1, 6525 ED Nijmegen, The Netherlands
Shiou-Min Wu, Dragana Č. Radenović, André J. A. van Roij,
Ivan Anton Garcia, and David H. Parkera?
Department of Molecular and Laser Physics, IMM, Radboud University Nijmegen, Toernooiveld 1,
6525 ED Nijmegen, The Netherlands
?Received 12 December 2006; accepted 18 January 2007; published online 6 March 2007?
Ultraviolet photodissociation of SH ?X2?, ??=2–7? and SD ?X2?, ??=3–7? has been studied at
288 and 291 nm, using the velocity map imaging technique to probe the angular and speed
distributions of the S?1D2? products. Photodissociation cross sections for the A2?+←X2????? and
2?←X2????? transitions have been obtained by ab initio calculations at the CASSCF-MRSDCI/
aug-cc-pV5Z level of theory. Both the experimental and theoretical results show that SH/SD
photodissociation from X2? ????7? proceeds via the repulsive wall of the A2?+state. The angular
distributions of S?1D2? indicate that the dissociation approaches the sudden recoil limit of the A2?+
state, yielding strongly polarized fragments. The S?1D2? atoms are predominantly produced with
total electronic angular momentum perpendicular to the recoil axis. © 2007 American Institute of
Physics. ?DOI: 10.1063/1.2646522?
I. INTRODUCTION
The mercapto radical ?SH? plays an important role in
atmospheric chemistry, particularly in the oxidation of H2S
and various other sulfur-containing compounds.1SH is also a
keyintermediatein combustion
containing fuels and has recently been observed in interstel-
lar space.2The photochemistry of this radical is therefore of
astrochemical importance.
The spectroscopy of the SH radical, and its isotopic ana-
log SD, has been extensively studied over the last decades,
both experimentally3–11and theoretically.12–17The electronic
configuration ofthe
X2?
?2??2?3??2?1??4?4??2?5??2?2??3, thus SH and SD are open-
shell systems. The first excited bound state, A2?+, is opti-
cally coupled to the ground X2? state and correlates with
H/D?2S1/2?+S?1D2? in the atomic limit. The A2?+state is
crossed by three repulsive curves ?4?−, 12?−, and
which lead to the ground state products H/D?2S1/2? and
S?3PJ?. The repulsive4?−, 12?−, and4? states are coupled
to the bound A2?+state via spin-orbit interactions, and the
A2?+state can therefore undergo predissociation through
these repulsive curves. Numerous studies have investigated
the predissociative character of the A2?+–X2? system, and
both radiative and predissociative lifetimes of SH/SD ?A2?+,
??=0–2? have been determined.13,17–22
Above the dissociation threshold of the H/D?2S1/2?
+S?1D2? channel, direct dissociation via the continuum re-
gion of the A2?+state becomes possible. The S?1D2? disso-
ciation limit can also be reached by optical excitation to the
processes ofsulfur-
groundstateis
?1??2
4??,
higher repulsive
curves of the doublet states correlating with the S?3PJ? and
S?1D2? atomic limits are shown in Fig. 1. Metastable S?1D2?
is more reactive than the S?3PJ? ground state atom, and the
S?1D2? dissociation channel is therefore important in various
chemical reactions, particularly in the astrochemical context.
The oxygen analog O?1D2? has already been observed in
cometarymatterandisprimarily
photodissociation.23Based on the relatively high cosmic
abundance of sulfur, it can be expected that S?1D2? is also
present in the interstellar medium, and photodissociation of
SH might play an important role in the production process.
The direct dissociation channels of SH have been studied
experimentallyintheultraviolet
ultraviolet ?VUV? region. Zhou et al.24studied the photodis-
sociation dynamics of jet-cooled SH ?X2?, ??=0–2? in the
wavelength region of 216–232 nm and concluded that the
UV photolysis of SH mainly proceeds via the repulsive 12?−
state. The observed S?3PJ? fine structure state distribution
indicated that the dissociation approaches the sudden recoil
limit of the 12?−state. However, they also noted that nona-
diabatic interactions among the repulsive4?−, 12?−, and4?
states influence the final S?3PJ? product distribution. No evi-
dence was found for the H?2S1/2?+S?1D2? channel in this
wavelength region, although the excitation energies are suf-
ficiently high to reach the S?1D2? limit. Continetti et al.25and
Hsu et al.26investigated the secondary photodissociation of
SH from photolysis of H2S at 193 nm. They concluded that
the observed S?3PJ? and S?1D2? products were most likely
coming from direct dissociation through the repulsive 12?−
and2? states, respectively. Chen et al.27recently studied the
VUV photodissociation of jet-cooled SH at 121 nm and
2? and 22? states. The potential energy
produced byOH
?UV?
andvacuum-
a?Author to whom correspondence should be addressed. Electronic mail:
parker@science.ru.nl
THE JOURNAL OF CHEMICAL PHYSICS 126, 094304 ?2007?
0021-9606/2007/126?9?/094304/8/$23.00© 2007 American Institute of Physics
126, 094304-1
Downloaded 07 Mar 2007 to 131.174.143.43. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

Page 2

found that the S?3PJ? product fine structure distribution is
significantly different from that in the UV region. They sug-
gested that the minor S?3PJ? product originates from initial
excitation to the2? or 22? state, which can couple nona-
diabatically with other repulsive states to produce S?3PJ?.
The observed production of S?1D2? was found to arise from
the repulsive 22? state. To our knowledge, no previous stud-
ies have detected S?1D2? via the continuum of the A2?+
state.
Theoretical studies of the direct S?3PJ? and S?1D2? pho-
todissociation channels of SH have been conducted by Lee et
al.28,29They predicted that the S?3PJ? and S?1D2? vector
properties show oscillatory variations due to interference
among the different dissociation pathways. However, these
findings seem to contradict the experimental results of Zhou
et al.,24who reported that the S?3PJ? fine structure state dis-
tributions and anisotropy parameters are nearly constant in
the energy range of 216–232 nm. Other theoretical work has
mainly focused on the predissociative behavior of the A2?+
state, and accurate ab initio calculations on the direct disso-
ciation channels of SH and SD are still limited in number.
This work aims to investigate the S?1D2? photodissocia-
tion channel of vibrationally excited SH/SD radicals at 288
and 291 nm, both experimentally and theoretically. Measure-
ments of the S?1D2? product velocities are compared with
cross sections for the A2?+←X2????? and2?←X2?????
transitions in order to identify the dominant dissociation
pathway. Additionally, the polarization of the sulfur frag-
ments is studied to provide a complete description of the
dissociation dynamics.
II. EXPERIMENT
The main concept of velocity map imaging30is to record
a two-dimensional projection of nascent species in velocity
space. The obtained image provides detailed information
about the speed and angular distributions of the photodisso-
ciation products. The experimental setup has been described
elsewhere,31,32and only a summary of the experimental de-
tails will be presented here.
A mixture of 25% D2S seeded in Xe was expanded into
a vacuum chamber through a pulsed Jordan valve ?10 Hz?,
where the SD radicals were produced by a pulsed electrical
discharge of ?3 kV ?10 ?s pulse? at the beginning of the
supersonic expansion. These discharge sources are known to
produce rotationally cold but vibrationally excited species.
Despite extensive flushing with D2S, it was not possible to
remove all H atoms from the surfaces of the gas handling
system. For this reason, significant and varying amounts of
SH were also formed in the discharge. The molecular beam
was collimated by a 1 mm skimmer, after which the SH/SD
molecules in the X2?3/2, J=3/2, f lambda doublet state
were focused using a 12 cm long hexapole with 6 mm diam-
eter rods, operating at ±10 kV. The molecular beam was then
crossed with a linearly polarized laser to induce both photo-
dissociation of the radicals and ionization of the S atoms.
The laser beam was generated by a frequency doubled dye
laser ?Quanta-Ray PDL-2?, pumped by a neodymium doped
yttrium aluminum garnet laser ?Quanta-Ray DCR-11,
10 Hz?. The laser wavelength was tuned at either 288.19 or
291.48 nm, for ?2+1? resonance enhanced multiphoton ion-
ization ?REMPI? detection of S?1D2? products via S?4p,1F3?
and S?4p,1P1?, respectively. The pulse energy of the UV
radiation was ?2 mJ. Subsequently, the S+fragments were
velocity mapped using an electrostatic lens, which consisted
of a repeller, extractor, and ground electrode. The ions were
projected onto a two-dimensional imaging detector consist-
ing of two microchannel plates and a phosphor screen. The
two-dimensional crushed image was recorded by a charge-
coupled device camera ?Pixelfly? and converted to a sliced
image of the original three-dimensional distribution by ap-
plying an inverse Abel transformation.33All measurements
were repeated several times.
III. CALCULATIONS
A. Photodissociation cross sections
In order to identify the pathway from which the ob-
served S?1D2? fragments arise, ??-dependent photodissocia-
tioncrosssectionswere
←X2????? and
electronic wave functions and potential energy curves of the
ground X2? state and excited A2?+and2? states were cal-
culated with the MOLPRO 2000 quantum chemistry package,34
using the augmented correlation-consistent polarized valence
quintuple-zeta ?aug-cc-pV5Z? basis set.35The potentials
were computed using the internally contracted multirefer-
ence singles and doubles configuration interaction ?MRS-
DCI? method36,37including the Davidson correction.38The
molecular orbitals were obtained from a complete active
space self-consistent field ?CASSCF? calculation,39,40where
the three lowest ? orbitals and the first ?xand ?yorbitals
were kept doubly occupied. The active space consisted of
three ? orbitals, one ?xorbital, and one ?yorbital.
The electronic transition dipole moments of the A2?+
←X2? and
calculatedforthe
A2?+
2?←X2????? transitions. The ab initio
2?←X2? transitions were calculated at the
FIG. 1. Photodissociation of SD ?X2?, ??=3? at 288 nm via the repulsive
wall of the A2?+state. The figure shows the calculated ab initio potentials,
A2?+←X2? transition dipole moment ?x?ea0?, bound X2? ???=3? wave
function, and continuum A2?+wave function. The 12?−and 22? poten-
tials are adapted from Ref. 17.
094304-2Janssen et al.J. Chem. Phys. 126, 094304 ?2007?
Downloaded 07 Mar 2007 to 131.174.143.43. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

Page 3

MRSDCI level in MOLPRO, with the molecular orbitals taken
from a state-averaged CASSCF calculation. The nuclear
wave functions and vibrational energies of the bound X2?
and A2?+states were obtained with the sinc-function dis-
crete variable representation method,41,42using a grid in the
range of 1–12a0with a step size of 0.02a0. The continuum
wave functions of the unbound region of the A2?+state and
the repulsive2? state were computed on the same grid with
the renormalized Numerov method.43The photodissociation
cross sections of the A2?+←X2????? and
transitions were then calculated using the relation44
???? =4?2???
3e2
2?←X2?????
???f?E???x?R??X2?x??????2,
?1?
where ?? is the energy of the absorbed photon, ? the fine
structure constant, e the elementary charge, ?f?E? the
energy-normalized dissociative nuclear wave function at to-
tal energy E, and ?x?R? the corresponding electronic transi-
tion dipole moment.
To evaluate the quality of the computed potentials,
the X2? vibrational frequencies of SH???=0–3? and
SD???=0–1? have been compared to the experimental data
from Ram et al.10and Pathak and Palmer,5respectively, and
were found to be accurate within 0.5%. Experimental fre-
quencies of the A2?+state are more difficult to determine
due to the occurrence of predissociation. Only the fundamen-
tal frequencies ??G1/2? have been reported for the A2?+
state of SH and SD,4which agree within 0.3% with the com-
puted values. Moreover, the calculated bond dissociation en-
ergy D0of the SH ground X2? state is 29 220 cm−1, which
is in excellent agreement with the recent experimental value
of 29 245±25 cm−1obtained by Zhou et al.24The computed
energy difference between S?3P? and S?1D? was found to be
8912.69 cm−1, which differs by 1.44% from the experimental
data.45The potentials of the A2?+and2? states have there-
fore been slightly shifted vertically to match the experimen-
tal S?3P?–S?1D? splitting.
B. Fragment polarization
The polarization of S?1D2? provides valuable insight into
the photodissociation dynamics and can be interpreted using
two different theoretical models. In the diabatic picture, the
dissociation is assumed to be very fast with respect to the
spin-orbit precession time, and the molecular wave function
is projected directly onto the atomic states. Conversely, the
adiabatic model assumes that the dissociation is much
slower, and spin-orbit coupling must be taken into account.
Both models are used in this study to calculate the S?1D2?
polarization arising from the A2?+←X2? direct dissocia-
tion channel, which is expected to be dominant in the UV
region. The results allow a qualitative analysis of the ob-
served S?1D2? alignment.
Experimentally, the polarization of the S?1D2? fragments
is reflected in the angular distribution of the velocity map
images. The observed ion distribution is the product of the
photofragment angular recoil distribution and the ionization
efficiency and can be written as an expansion of ordinary
Legendre polynomials Pl,46,47
Iion??? ? ?
l=0,2,4,6
clPl?cos ??,
?2?
where ? is the recoil angle with respect to the polarization
axis of the laser. In the case of a purely perpendicular tran-
sition, for which the anisotropy parameter ? equals −1, the
expansion coefficients are given by:47,48
c0= 1 −1
5
?0
?0
?2?
?0?I2,
c2=5
7
?0
?0
?2?
?0?I2−2
7
?0
?0
?4?
?0?I4− 1,
?3?
c4= −18
35
?0
?0
?2?
?0?I2+57
77
?0
?0
?4?
?0?I4,
c6= −
5
11
?0
?0
?4?
?0?I4,
where Ikare the relative geometrical factors of the ionization
scheme and ?0
matrix ?state multipoles? for the sulfur atom in the body-
fixed frame. The factors Ikcan be calculated using the theory
by Mo et al.46,49It follows that I0:I2:I4=1:4?70/49:
−3?14/98 for the REMPI scheme via S?1F3? at 288 nm and
I0:I2:I4=1:−?70/14:−2?14/7 for the S?1P1? resonance at
291 nm. The state multipoles ?0
Eq. ?24? of Ref. 47,
?k?=?
?S
?k?are the irreducible components of the density
?k?can be obtained using
?0
?− 1?JS−?S?JS,?S;JS,− ?S?k,0???S;?S,
?4?
where the symbol ?JS,?S;JS,−?S?k,0? is a Clebsch-Gordan
coefficient. The quantum number ?Sdenotes the projection
of the total angular momentum of the sulfur atom JS=2 with
respect to the internuclear ?recoil? axis. The term ??S;?Srep-
resents the ?Sstate populations, i.e., the diagonal elements
of the density matrix for sulfur. Note that the off-diagonal
elements of this density matrix are not taken into account,
which is a direct consequence of ignoring the possible effects
of coherent excitation. The populations in the different ?S
states effectively determine the polarization of the S?1D2?
atom, and it is thus necessary to evaluate the density matrix
explicitly. In this study, both the diabatic and adiabatic mod-
els are employed to predict which ?Sstates are populated.
In the diabatic model, the molecular A2?+state is de-
scribed by the Hunds case ?a? electronic quantum numbers
?, S, and ?, denoting the body-fixed electronic orbital angu-
lar momentum projection, total electronic spin, and spin pro-
jection, respectively. The atomic fragments are described by
the corresponding atomic quantum numbers ?i, Si, and ?i,
where i labels the particular atom. The quantum numbers Li
refer to the orbital angular momentum of the atoms. As spin-
orbit recoupling is neglected in this model, ?=?S+?His
conserved during the photodissociation process. Given that
?=0 for the A2?+state and ?H=0 for the H?2S1/2? atom, it
follows that the S?1D2? fragment is produced in the ?S
=?S=0 state. This reduces the summation in Eq. ?4? to a
single term, and the expansion coefficients clcan be readily
094304-3
Photodissociation of SH and SD: S?1D2? channelJ. Chem. Phys. 126, 094304 ?2007?
Downloaded 07 Mar 2007 to 131.174.143.43. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

Page 4

obtained from Eq. ?3?. To compare these results with the
experimental data, the expansion coefficients are expressed
in terms of normalized Legendre moments c ˜l=cl/c0.47,48The
values of c ˜lare given in Table I for both detection schemes.
In the spin-orbit adiabatic limit, ?=?+? is assumed to
be conserved during dissociation. The A2?+state is labeled
by the quantum number ???=1/2 and correlates to ??S?H? in
the long range. The higher lying 22? state also has an ???
=1/2 fine structure component but correlates to a different
??S?H? state. In order to construct the adiabatic correlation
diagram, the energy ordering of the ??S?H?=?0,1/2? and
?1,−1/2? states must be evaluated at long range. As the spin-
orbit interaction is zero for both atoms, we assume that spin-
orbit coupling is negligible at large internuclear distances.
Moreover, the long range quadrupole-quadrupole interaction
is zero, since the hydrogen atom is produced in an S state.
Consequently, the leading term that lifts the degeneracy of
the ??S?H? states is the dispersion interaction. This is also
the case in, e.g., the He?1S?–Sc?2D? system, which has been
worked out explicitly by Chu et al.50Following their nota-
tion, the dispersion interaction between H?2S1/2? and S?1D2?
as a function of the internuclear distance R is written as
V??R? ? −C6?L,??
R6
,
?5?
where L=LS=2. The quantum number ?=?S=0 corresponds
to the ??S=0,?H=1/2? state and ?=?S=1 to the ??S
=1,?H=−1/2? state. Hence, we must determine the energy
ordering of the Born-Oppenheimer potentials V0?R?=V??R?
and V1?R?=V??R? at long range. The dispersion coefficients
C6?L,?? are given by
C6?L,?? = C6,0?L? −
3?2− L?L + 1?
?2L − 1??2L + 3?C6,2?L?,
?6?
where C6,0?L? and C6,2?L? denote the scalar and ?rank 2?
tensor components of the dispersion interaction, respectively.
From this equation, it follows that the energy ordering of the
? states is determined by ?2C6,2?L?. The sign of C6,2?L? can
be evaluated using the following expression:50
C6,2?L? = −3?2L + 3?
2?L?
0
?
?2?L;i??? ¯H?i??d?,
?7?
where ?2?L;i?? is the tensor polarizability of the S?1D2?
atom at imaginary frequency i? and ? ¯H?i?? is the dynamic
polarizability of the H?2S1/2? atom. The factor ? ¯H?i?? is posi-
tive for all angular frequencies ? ?Ref. 51? and takes its
maximum value at ?=0. Although we did not find ?2?L;i??
in the literature, we assume that the sign of the integral is
determined by the sign at ?=0, which also holds for the
He?1S?–Sc?2D? and He?1S?–Ti?3F? systems.50The sign of
?2?L?=?2?L,?=0? can be derived using Eq. ?11? of Ref. 50,
where the polarizability anisotropy ???L,?;i?? is defined.
In the case of ?=0 and ?=0, this equation may be rewritten
as
?2?L? = −2
3???L,0?L?2L − 1?
L?L + 1?.
?8?
The quantity ???L,0? has been calculated by Medved et
al.,52who reported a negative value of −6.43 a.u. for S?1D2?.
Consequently, the tensor polarizability ?2?L? is positive,
C6,2?L? is negative, and the total dispersion interaction de-
pends linearly on −?2. The energy ordering in the long range
is thus given by V2?R??V1?R??V0?R?, which leads to the
FIG. 2. Adiabatic correlation diagram for the S?1D2?+H?2S1/2? photodisso-
ciation channel of SH. The quantum number ? denotes the projection of the
total electronic angular momentum with respect to the internuclear axis and
correlates to ?S+?Hin the long range.
TABLE I. Simulation parameters for the TKER and angular distributions of the S?1D2? velocity map images,
obtained at 288 and 291 nm using REMPI detection via S?1F3? and S?1P1?, respectively. The parameter T
denotes the vibrational temperature, SH/SD is the total signal ratio between SH and SD, ?v is the velocity
resolution, and c ˜lare the normalized Legendre moments. The calculated Legendre moments for the diabatic and
adiabatic models are also given.
T ?K?
SH/SD
?v ?m/s?
c ˜2
c ˜4
c ˜6
S?1F3?←S?1D2?
Image ?a?
Image ?b?
Diabatic
Adiabatic
2300
2700
¯
¯
1.7
1.4
¯
¯
35
43
¯
¯
−0.96
−0.92
−1.32
−1.23
0.10
0.13
0.24
0.28
0.08
0.02
0.07
−0.05
S?1P1?←S?1D2?
Image ?c?
Image ?d?
Diabatic
Adiabatic
2500
2200
¯
¯
1.2
0.2
¯
¯
38
35
¯
¯
0.24
−0.48
0
−1.15
−1.38
−1.04
−1.91
0.71
0.46
0.64
0.91
−0.56
094304-4Janssen et al.J. Chem. Phys. 126, 094304 ?2007?
Downloaded 07 Mar 2007 to 131.174.143.43. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

Page 5

correlation diagram shown in Fig. 2. Note that the energy
ordering of the molecular A2?+, 22?, and2? states is based
on the ordering in the Franck-Condon region.
Figure 2 shows that the A2?+state correlates adiabati-
cally with the ??S?=1 component of S?1D2?, whereas the
diabatic model predicts ?S=0. The normalized Legendre
moments for the adiabatic limit are given in Table I.
IV. RESULTS AND DISCUSSION
Figure 1 illustrates the photodissociation of SD ?X2?,
??=3? at 288 nm based on the computed potentials and wave
functions. The largest Franck-Condon overlap between the
ground state and the continuum region of the A2?+state is
found at the inner wall of the potential. The repulsive
state is hardly accessible at this wavelength.
2?
A. Photodissociation cross sections
The photodissociation cross sections of SH/SD for the
A2?+←X2????? and2?←X2????? transitions, both yield-
ing S?1D2?, are presented in Fig. 3. Numerical values are
available from the authors upon request. It can be seen that
the width of the cross section peaks is significantly larger for
SH than for SD. This broadening can be qualitatively under-
stood based on the reflection principle, as the vibrational
states of SH are more delocalized than the states of SD.
At the S?1D2? REMPI wavelengths of 288 and 291 nm,
it is evident from Fig. 3 that S?1D2? can only be produced by
direct dissociation of vibrationally excited SH/SD via the
repulsive part of the A2?+state, whereas the
transition is dominant in the VUV region. It is expected that
the 22? state becomes important at even higher energies.
Note that the excitation wavelengths of 288 and 291 nm are
insufficient to produce S?1D2? from SH ?X2?, ??=0–1? and
SD ?X2?, ??=0–2?. The photodissociation cross sections
are consistent with the work of Continetti et al.25and Hsu et
al.,26who both reported that the S?1D2?+H?2S1/2? dissocia-
tion channel at 193 nm proceeds via the2? state. The recent
work of Chen et al.27showed that the S?1D2? product of SH
photodissociation at 121 nm mainly arises from the repulsive
22? state rather than the2? state. Given that the computed
cross sections for the2?←X2? transition are nearly zero at
this high energy, the 22? state is thus expected to be domi-
nant. In the UV region of 216–232 nm, Zhou et al.24found
no evidence for the S?1D2?+H?2S1/2? dissociation channel of
SH???=0–2?, which is also consistent with the calculated
cross sections. In this wavelength region, the cross sections
for both the A2?+←X2? and
negligible for the lowest vibrational states.
2?←X2?
2?←X2? transitions are
B. TKER distributions
The experimental S+velocity map images are shown in
Fig. 4. The central spots in images 4?a?–4?c? originate from
S?1D2? produced directly by the electrical discharge, yielding
fragments with zero velocity in the plane of the detector.
Image 4?d? was obtained under slightly different experimen-
tal conditions, producing less S?1D2? in the discharge. The
FIG. 3. Calculated ??-dependent SH/SD photodissociation cross sections for
the S?1D2? channel in the wavelength region of 130–350 nm. The dashed
lines correspond to the A2?+←X2????? transition and the solid lines to the
2?←X2????? transition. The cross sections for X2????=1–8? are each
plotted with an offset of 0.1?10−18cm2for clarity.
FIG. 4. Raw velocity map images of S?1D2? from SH/SD photodissociation
at ??a? and ?b?? 288 nm and ??c? and ?d?? 291 nm. The S?1D2? fragments
were detected by ?2+1? REMPI via S?1F3? and S?1P1?, respectively. The
signal intensity is indicated in a gray scale, where the darkest area corre-
sponds to the highest signal. The laser polarization is along the vertical axis
in the figure.
094304-5
Photodissociation of SH and SD: S?1D2? channel J. Chem. Phys. 126, 094304 ?2007?
Downloaded 07 Mar 2007 to 131.174.143.43. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp