An initial investigation of S and SH with angle resolved photoelectron spectroscopy using synchrotron radiation

Article abstract of DOI:10.1016/j.chemphys.2003.11.011

An initial investigation of the photoionization behaviour of S atoms and SH radicals has been made in the gas-phase using angle resolved photoelectron spectroscopy (PES) with radiation from the Elettra synchrotron as the photon source. Angular distributions and relative photoionization cross-sections have been measured for S and SH as a function of photon energy from threshold (～10.3 eV) to 21.64 eV. The ionizations considered are S⁺( ⁴S)←S(³P), S⁺(²D)←S( ³P) and S⁺(²P)←S(³P), and SH⁺(X³Σ^-, v⁺=0)←SH(X ²Π, v″=0), SH⁺(a¹Δ, v ⁺=0)←SH(X²Π, v″=0) and SH⁺(b ¹Σ⁺, v⁺=0)←SH(X²Π, v″=0). This study of S and SH follows a recent investigation on atomic oxygen where the feasibility of angle resolved PES measurements on reactive intermediates at the Elettra synchrotron source was demonstrated, and further studies on small molecular radicals were proposed.

Full text of DOI:10.1016/j.chemphys.2003.11.011

Chemical Physics 298 (2004) 213–222
www.elsevier.com/locate/chemphys
An initial investigation of S and SH with angle resolved
photoelectron spectroscopy using synchrotron radiation
a
L. Zuin , F. Innocenti , M.L. Costa , A.A. Dias , A. Morris ,
a
b
b
a
b
c
d
A.C.S. Paiva , S. Stranges , J.B. West , J.M. Dyke
a,
*
a
Department of Chemistry, University of Southampton, Highfield, Southampton SO17 1BJ, UK
CEFITEC Department of Physics, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal
b
c
Dipartimento di Chimica, Universita di Roma ‘‘La Sapienza’’, I-001285 Rome, Italy and TASC-INFM Laboratory, Basovizza, 34012 Trieste, Italy
d
CLRC Daresbury Laboratory, Daresbury, Warrington WA4 4AD, UK
Received 20 August 2003; accepted 12 November 2003
Abstract
An initial investigation of the photoionization behaviour of S atoms and SH radicals has been made in the gas-phase using angle
resolved photoelectron spectroscopy (PES) with radiation from the Elettra synchrotron as the photon source. Angular distributions
and relative photoionization cross-sections have been measured for S and SH as a function of photon energy from threshold (ꢀ10.3
þ4
3
þ2
þ
3
þ2
3
þ
3
ꢁ
eV) to 21.64 eV. The ionizations considered are S ( S) S( P), S ( D) S( P) and S ( P) S( P), and SH (X R ,
v
þ
2
00
þ
1
þ
2
00
1
þ
þ
2
00
¼0) SH(X P, v ¼0), SH (a D, v ¼0) SH(X P, v ¼0) and SH (b R , v ¼0) SH(X P, v ¼0).
This study of S and SH follows a recent investigation on atomic oxygen where the feasibility of angle resolved PES measurements
on reactive intermediates at the Elettra synchrotron source was demonstrated, and further studies on small molecular radicals were
proposed.
Ó 2003 Elsevier B.V. All rights reserved.
1
. Introduction
The ground state configuration of S atoms is
3
2
2
6
2
4
ꢁ1
1
s 2s 2p 3s 3p , P. Three ionic states arise on (3p)
4
2
2
In recent experiments at the Elettra synchrotron, we
ionization from this state: the S, D and P at ioniza-
tion energies of 10.35, 12.21 and 13.40 eV, respectively
[5].
have studied atomic oxygen in the gas-phase using angle
resolved photoelectron spectroscopy [1]. This work ex-
tends these measurements to S and SH.
There has been considerable theoretical work carried
out on the photoionization cross-section of S atoms over
a wide photon energy range. Conneelly et al. [6] com-
puted photoionization cross-sections for atomic systems
with incomplete outer 3p subshells using the continuum
Hartree–Fock approximation, and Dill et al. [7] derived
expressions for the angular distribution of photoelec-
trons produced by ionization of open shell atoms in-
cluding sulphur. The asymmetry parameter, b, as
function of photon energy was calculated for each of the
In the extreme ultraviolet wavelength region, ab-
sorption, photoionization and autoionization of sulphur
atoms are of fundamental interest for understanding the
spectroscopy and photoionization dynamics of open-
shell atoms. Moreover, due to its relatively high solar
abundance, sulphur plays an important role in deter-
mining solar and stellar opacities [2,3] and interaction of
S atoms with ultraviolet radiation is important in the
solar atmosphere [4].
ꢁ
1
three ionic channels resulting from the (3p) ionization
of sulphur atoms using both Hartree–Fock and Her-
man–Skillman wavefunctions for the atom, ion and free
electron from threshold up to 60 eV.
*
Corresponding author. Tel.: +44-23-8059-3590; fax: +44-23-8059-
Later Manson et al. [8] presented results of calcula-
tions of the photoionization cross-sections and angular
3781.
E-mail address: jmdyke@soton.ac.uk (J.M. Dyke).
0
301-0104/$ - see front matter Ó 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.chemphys.2003.11.011

214
L. Zuin et al. / Chemical Physics 298 (2004) 213–222
distributions of the outer p subshell of chalcogen and
halogen atoms from 13 to 100 eV photon energy, using
the continuum Hartree–Fock approximation. In the
theoretical work on S atoms [7,8], the general shape
of the cross- section for each of the three ionic channels
shows a clearly observable Cooper minimum around 40
eV. The calculated asymmetry parameters as a function
of photon energy have a small value near threshold, rise
rapidly to a maximum then fall to a minimum with
negative b values at the Cooper minimum of the cross-
section; they subsequently increase smoothly (the S 3p
orbital is expected to show a Cooper minimum since the
number of nodes in the atomic wavefunction, n ꢁl ꢁ1,
gated from the near ultraviolet into the vacuum ultra-
violet region and several electronic transitions have been
identified [16,17].
2
The adiabatic ionization energies of SH (X P) to the
first five ionic states have been established from HeI
vacuum ultraviolet (VUV) photoelectron spectroscopy
measurements to an accuracy of ꢂ0.01 eV [5]. Recently
a more accurate determination of the ionization energies
3
ꢁ22
1
2
to the X R
have been achieved by the pulsed field ionization zero
and a D ionic state [23] from SH X P
electron kinetic energy method.
The ground state electronic configuration of the SH
radical is
ꢁ
1
is greater than zero). The (3p) photoionization cross-
section of S atoms has also been calculated at a number
of photon energies by Yeh and Lindau [9] using Har-
tree–Fock–Slater calculations.
2
2
2
4
2
2
3
1
r 2r 3r 1p 4r 5r 2p
Photoionization from the valence 2p molecular or-
Experimentally, a number of studies have been made
on S atoms [10–13]. Tondello [10] measured and analy-
sed the absorption spectrum of S in the vacuum ultra-
3
ꢁ
1
1
þ
bital produces the X R , a D and b R ionic states at
adiabatic ionization energies (AIEs), equal to the verti-
cal ionization energies (VIEs), of 10.42, 11.64 and 12.76
eV, respectively [5,22,23]. The potential energy curves of
these ionic states are very similar to that of the ground
state of the neutral because of the non-bonding S 3p
character of the 2p orbital. The removal of one electron
ꢀ
violet region 1830–900 A (6.77–13.77 eV), below and
above the ionization threshold. An attempt to derive the
absolute value of the photoionization cross section from
ꢀ
the ground state in the 1197–940 A region (10.36–13.19
eV) was made in this work. Modifications and additions
to this spectrum have been made using a similar tech-
nique by Sarma and Joshi [11] and, more recently, by
Joshi et al. [12]. Subsequently, Gibson et al. [13] studied
the photoionization spectrum of atomic sulphur using
photoionization mass spectrometry from the ionization
threshold, 10.35 to 13.40 eV.
The SH radical is a molecule of environmental im-
portance which plays a role in the ultraviolet photo-
chemistry of several small sulphur-containing
compounds which are released into the earthÕs atmo-
sphere from natural and anthropogenic sources [14]. For
example, hydrogen sulphide, carbonyl sulphide and
carbon disulphide are released from the earth and its
oceans as a consequence of microbiological processes.
They react with the hydroxyl radical in the atmosphere
to yield SH, which is then oxidized by O2 and O3 to SO2.
Although SH has been studied with various spectro-
scopic techniques [15–18] and theoretical methods
3
1
from the 5r shell gives rise to the A P and c P ionic
states at AIEs of 14.11 and 15.69 eV, respectively [5].
þ
3
2
þ
1
2
The SH (A P) SH(X P) and SH (c P) SH(X P)
bands are broad and vibrationally resolved, with the
AIE equal to the VIE for both bands. AIE values for
these bands are 14.11 and 15.69 eV, respectively [5].
Until now no attempts have been made to measure
asymmetry parameters (b) and photoionization cross-
sections (r) of either S or SH. The purpose of this
present work was to measure angular distribution pa-
rameters and relative photoionization cross-sections for
þ4
3
þ2
3
þ2
2
3
the S ( S) S( P), S ( D) S( P) and S ( P) S( P)
00
þ
3
ꢁ
þ
bands and the SH (X R , v ¼0) SH(X P, v ¼0),
þ
1
þ
2
00
þ
1
þ
SH (a D, v ¼0) SH(X P, v ¼0) and SH (b R ,
þ
2
00
v ¼0) SH(X P, v ¼0) bands as a function of
photon energy from threshold (ꢀ10.3 eV) to 21.64 eV
with the Elettra synchrotron source. This should allow
comparison to be made with available calculations.
[
19,20], information on its electronic states is still
somewhat limited. A multiphoton ionization study of
the SH radical in the single photon energy range 4.68-
2. Experimental
3
.91 eV has been performed by Ashfold et al. [15] The
observation and partial characterization of three Ryd-
berg states of SD and SH, accessed via 2-photon reso-
nances from the ground state, was achieved in this work.
More recently, thirteen Rydberg states converging to the
second and third ionic state have been observed and
characterised by two-photon resonance enhanced mul-
tiphoton ionization spectroscopy and multiphoton ion-
ization photoelectron spectroscopy [21]. Also, the
electronic absorption spectrum of SH has been investi-
The experiments reported here were undertaken on
beam-line 4.2 R of the Circularly Polarized Beamline at
the Elettra Synchrotron Light Source (Trieste). From
the synchrotron source, the radiation passes to the
electron spectrometer via a monochromator, which
covers the photon energy range 5–35 eV. The photo-
electron spectrometer used in this work has been spe-
cifically designed to study reactive intermediates with
photoelectron spectroscopy using synchrotron radiation

L. Zuin et al. / Chemical Physics 298 (2004) 213–222
215
and has been described in detail previously [24,25].
Briefly, it consists of an ionization chamber, a three el-
ement lens assembly, a 10 cm mean radius hemispherical
analyser and a channeltron detector. It can be rotated
around the axis of linear polarization of the photon
beam in the range h ¼0° to h ¼73:5°, allowing the
angular distributions of photoelectrons to be studied.
Photoelectron (PE) spectra were recorded in the
constant pass energy mode, by scanning the voltage on a
lens which accelerates (or decelerates) the photoelec-
trons before they enter the analyser.
where R ¼I
0
=I54°440 is the ratio of the experimental in-
tensities at these two angles, after applying the above
corrections.
3. Results and discussion
Test experiments were carried out in Southampton in
order to determine the optimum pressures and mixing
distance above the photon beam, to be used in the ex-
periments at Elettra, which maximise the intensities of
the second and third photoelectron bands of SH at 11.64
and 12.76 eV AIE, corresponding to the ionizations
In this work, S atoms and the SH radicals were
produced by the following consecutive reactions:
þ
1
þ
2
00
þ
1
þ
SH (a D, v ¼0) SH(X P, v ¼0) and SH (b R ,
F þH
2
S !SH þHF
þ
2
00
v ¼0) SH(X P, v ¼0), and the second and third
photoelectron bands of S at 12.21 and 13.40 eV IE,
þ2
F þHS !S þHF
3
corresponding to the ionizations S ( D) S( P) and
þ2
3
S ( P) S( P) [5]. The following conditions were found
to be the optimum for production of both S and SH: a
mixing distance above the photon beam of 2 cm, with
the inner-inlet system used for H2S longer than the outer
inlet system by about 2 mm, and the partial pressures of
H2S and F2(5%)/He measured on an ionization gauge
positioned on the wall of the ionization chamber of
The room temperature rate constants for the above
ꢁ
10
ꢁ10
3
reactions are 1.33 ꢃ10
and 2.01 ꢃ10
ꢁ1
s , respectively [26]. To obtain a high F-atom
cm mole-
ꢁ
1
cule
yield, fluorine atoms were produced by flowing 5% F in
2
helium through a microwave discharge at 2.45 GHz in
the side arm of a glass inlet system [27], with an inner
inlet system which is the inlet used for the target mole-
cules (in this case H2S). In order to avoid attack of the
glass system by the fluorine atoms produced by the
discharge, the inner surface of the tube was coated with
a thin layer of teflon, while the discharge cavity was
positioned on an alumina section of the inlet system to
prevent melting of the teflon due to heating from the
discharge [27]. The intensities of the bands in the ex-
perimental PE spectra were normalized by the photon
flux and then by the transmission correction of the
spectrometer. The transmission function of the spec-
trometer, for electrons of different initial electron kinetic
energy, was determined by measuring the intensity of
2
.4 ꢃ10ꢁ and 4.6 ꢃ10 mbar, respectively. The PE
7
ꢁ6
spectra recorded at a photon energy of hm ¼21:64 eV
with synchrotron radiation and at two different angles
with respect to the polarization axis of the photon
0
source (h ¼0° and h ¼54°44 ) of the products of the
F + H2S reaction are shown in Figs. 1(a) and (b) with the
assignment of the major features indicated. These
spectra show sharp bands at 11.64 and 12.76 eV, which
are the second and third bands of SH, and sharp bands
at 12.21 and 13.40 eV which are the second and third
bands of S atoms. The sharp band at ꢄ10.5 eV arises
from overlap of the first band of S atoms at 10.35 eV,
and the first band of SH at 10.42 eV as well as a small
contribution from the first band of residual H2S at 10.48
eV [5]. Bands associated with F and F2 were also ob-
served in the spectra [30,31], as well as a small contri-
bution from SF2 which is a secondary reaction product
ꢁ
1
the helium (1s) photoelectron band at different photon
ꢁ
1
energies. The intensity of the helium (1s) band was
measured in the VUV photon energy range used and
then it was normalised using the absolute photoioniza-
tion cross-section of helium in this range [28]; finally the
normalized intensity was plotted as function of the
electron kinetic energy. This curve was used to correct
the experimental PE spectra for the transmission of the
spectrometer.
[32] (see Figs. 1 and 2). The vibrationally resolved band
of SH at 14.11 eV AIE is the fourth band of SH (the
þ
3
2
SH (a P) SH(X P) ionization). Unfortunately, the
þ
1
2
fifth band of SH (the SH (c P) SH(X P) ionization)
AIE 15.69 eV, was not observed because it was to weak
and overlapped with the first band of F2.
The degree of linear polarisation (P ¼1) of the
radiation was well established [29]. The asymmetry
parameter (b) was measured for S and SH radicals at
selected photon energies and over a photon energy
range, by recording PE spectra at two different angles
In Fig. 2(b), an expanded version of the composite
band at ꢄ10.5 eV is shown. It clearly shows two max-
ima. Deconvolution of this band into two Gaussians
was performed and the fitted Gaussians were centred at
0
0
(
0° and 54°44 ) at each photon energy (at 54°44 the
1
0.36 ꢂ0.02 and 10.45 ꢂ0.03 eV. These energies are in
measured intensity is proportional to the total pho-
toionization cross-section, independent of b); the b
parameter was then calculated from
þ4
agreement with the known energies of the S ( S)
ꢁ
3
þ
3
2
S( P) and SH (X R ) SH(X P) ionizations at 10.35
and 10.42 eV, respectively. The small contribution of
þ
2
1
b ¼ðR ꢁ1Þ;
ð1Þ
the H2S (X B1) H2S(X A1) ionization at 10.48 eV is

216
L. Zuin et al. / Chemical Physics 298 (2004) 213–222
F
SF2
S; SH
H S
2
HF
F2
S
SH
S
SH
SH
10
11
12
13
14
15
16
17
10
11
12
13
14
(
a)
Ionization Energy/eV
(a)
Ionization Energy/eV
S; SH
θ = 0
θ = 54 44'
S
SH
S
SH
SH
10
11
12
13
14
10.2
10.3
10.4
10.5
10.6
(
b)
Ionization Energy/eV
(b)
Ionization energy/eV
Fig. 1. Photoelectron spectra recorded at hm ¼21:64 eV for the reac-
tion F + H2S at a mixing distance of 2 cm above the photon beam (a) at
an angle (h) with respect to the polarization axis of the photon source
of 0°, and over the ionization energy range 10.0–17.5 eV, and (b) at
Fig. 2. (a) Photoelectron spectrum recorded at hm ¼21:64 eV at h ¼0°
for the reaction F + H S at a mixing distance of 2 cm above the photon
2
beam. Estimated contributions from SF (first and second bands) and
2
H S (second band) are shown. (The asymmetry parameter of the first
2
0
two angles h ¼0° and 54°44 and over the ionization energy range
vibrational component of the 4th band of SH at 14.11 eV IE has been
10.0–14.7 eV. (The photon energy was set to 21.22 eV at Elettra but,
measured as (1.04 ꢂ0.39) at hm ¼21:64 eV.) (b) Deconvolution of the
band shown in (a) at ꢄ10.4 eV ionization energy into two Gaussian
components. The band centred at 10.36 eV is the first band of S atoms
after calibration of the monochromator, the actual photon energy was
1.64 eV.)
2
þ
4
3
(
S ( S) S( P)) whereas the band at 10.45 eV contains contributions
þ
3
ꢁ
2
from the first band of SH (SH (X R ) SH(X P)) and from the first
band of H2S (see text for further details).
included in the band centred at 10.45 eV. This is con-
sistent with a slightly larger full-width-at-half-maximum
(
FWHM) of the second fitted Gaussian compared to the
tion energy side of the sharp band at ꢄ10.5 eV (1st AIE
first Gaussian. The values obtained from the fit are
FWHM1 ¼72 ꢂ3 meV and FWHM2 ¼95 ꢂ3 meV for
the first and second Gaussian, respectively.
of SF 10.08 eV) [32] (see Fig. 2(a)). The spectra ob-
2
tained are in good agreement with those previously re-
corded in Southampton by Dunlavey et al. [5] with a Hel
radiation source.
The ionization energy region 9.600–15.200 eV was
recorded using 23 different photon energies from 13.205
to 21.640 eV. A step size of 0.20 eV was used from
hm ¼13:205 eV to hm ¼16:420 eV and a step size of 0.40
eV was used from hm ¼16:420 eV to hm ¼18:494 eV; PE
spectra were also recorded at hm ¼21:640 eV. All spec-
The broad band on the low energy side of the
þ2
3
S ( P) S( P) ionization at 13.40 eV ionization energy
in Fig. 2(a) is mainly due to the presence of un-reacted
H S, along with SF produced by secondary reactions.
2
2
Confirmation of the presence of SF can be easily ob-
2
tained from the observation of the first two vibrational
components of the first band of SF2 on the low ioniza-

L. Zuin et al. / Chemical Physics 298 (2004) 213–222
217
0
tra were recorded at two angles, h ¼0° at h ¼54°44 ,
2.5
2.0
1.5
1.0
0.5
with respect to the polarization axis of the photon
source. Due to significant variations in the intensities of
the 12.21 eV band (second band of S) and the 11.64 eV
band (second band of SH) with photon energy, the
ionization energy region 11.500–12.400 eV was re-ex-
amined by recording a new series of 21 PE spectra with
photon energies from hm ¼13:205 eV to hm ¼13:411 eV,
with a smaller step size of 0.01 eV. A strong sharp band
was observed in the PE spectra recorded at photon en-
ergies above 12.5 eV, due to ionization of helium by
second order radiation. In this photon energy region
second order radiation can ionize helium (IE ¼24.58
eV [33]) with consequent emission of electrons. This
+
4
3
0
.0
(a) S ( S)
S( P)
-
-
0.5
1.0
2.5
2.0
1.5
1.0
0.5
0.0
+
2
3
ꢁ
1
(b) S ( D)
S( P)
helium (1s)
band was used as internal check for
-
-
0.5
1.0
the correction factor applied for the evaluation of the
asymmetry parameter for the observed bands of S and
SH. Indeed, with the procedure used, b was always
obtained as 2.00, within experimental error, for this
2.5
2.0
1.5
1.0
0.5
ꢁ
1
helium (1s) band. In practice, for the measurements of
angular distribution it is important to make sure that
the efficiency of the analyser does not change when the
spectrometer is rotated. The biggest contribution to the
change of the efficiency of the analyser comes from small
misalignments between the axis of rotation and the di-
rection of the photon beam. This contribution is mini-
mised by adjusting, after each rotation of the
spectrometer, the direction of the photon beam in such a
way as to maximise the signal coming from the detector
and the current measured by the photodiode on the
opposite side of the ionization chamber of the spec-
trometer from the direction of the photon beam. How-
ever, to quantify this contribution for all the angle
resolved experiments a correction factor is evaluated at
each set of experiments by measuring the intensity of
0.0
+
2
3
-0.5
-1.0
(c) S ( P)
S( P)
13
14
15
16
17
18
19 20
21
Photon Energy/eV
Fig. 3. Plots of the asymmetry parameter (b) as a function of photon
energy over the range hm ¼13:0–21.0 eV, for the first three photo-
electron bands of S atoms. (b) includes b values for the first band of
H2S measured at different photon energies in [34], denoted with the
symbol (*) in the figure.
then a slow increase. The first bands of S, SH and H2S
are expected to be very similar in their b and r plots as
they all correspond to ionization from a S 3p lone-pair
orbital. Comparisons of the measured asymmetry pa-
rameters as a function of photon energy for the three
0
bands at h ¼0° and at h ¼54°44 , usually at 21.22 eV
photon energy, from a sample (usually Ar) where the
asymmetry parameter is well established. From the
comparisons between the calculated asymmetry param-
eter and the known asymmetry parameter, the correc-
tion factor was determined. Then, at each photon
energy, the asymmetry parameter can be calculated for
an S or SH band using Eq. (1) with this correction
factor, and relative photoionization cross-sections can
be evaluated by integrating the area of the bands at
ꢁ
1
ionization channels resulting from the (3p)
photo-
ionization of S derived from this work with results of the
angular distribution Hartree–Fock calculations of Dill
et al. [7] have been made over the photon energy range
10–22 eV. These calculations [7] used LS coupling and
were based on an angular momentum transfer expan-
sion for the differential photoionization cross-section.
Dill et al. [7] showed that the three ionic channels re-
0
h ¼54°44 . This has been carried out to give the plots
shown in Figs. 3–8.
3
Inspection of existing angular distribution and rela-
tive cross-section (r) measurements for H2S [34–36],
indicates that for the S photoelectron bands arising from
sulting from photoionization of S( P) should show
significantly different behaviour from each other when
plotted as function of photon energy, in contrast to
identical behaviour expected in the Cooper–Zare model
[37]. In Fig. 3, the three different angular distribution
plots are shown as a function of photon energy and
comparison with the results of Dill et al. [7] is presented
in Fig. 4. As can be seen in Fig. 4, the experimental re-
sults obtained in this work are in reasonable agreement
ꢁ
1
a (3p) ionization, the angular distribution parameter
b) plotted as function of photon energy is expected to
(
show an initial increase, a maximum at ꢄ25 eV and a
minimum at ꢄ40 eV, the Cooper minimum, whereas the
r-plot is expected to show a regular decrease from
threshold, the shallow Cooper minimum at ꢄ40 eV and

218
L. Zuin et al. / Chemical Physics 298 (2004) 213–222
2.5
2.0
1.5
1.0
0.5
0.0
+
4
3
+
2
3
S ( S)
S( P)
S ( D)
S( P)
2.5
2.0
1.5
1.0
0.5
0.0
experimental data
calculations ref.(7)
experimental data
calculations ref.(7)
-
-
0.5
1.0
-
-
0.5
1.0
1
0
20
30
40
50
60
70
80
90 100
10
20
30
40
50
60
70
80
90 100
Photon Energy/eV
Photon Energy/eV
2.5
2.0
1.5
1.0
0.5
0.0
2.5
2.0
1.5
1.0
0.5
0.0
-
0.5
-0.5
10
11
12
13
14
15
16
17
18
19
20
12 13 14 15 16 17 18 19 20 21 22 23
Photon Energy/eV
(
a)
Photon Energy/eV
(b)
+
2
3
S ( P)
S( P)
experimental data
calculations ref.(7)
2.0
1.5
1.0
0.5
0.0
-
-
0.5
1.0
10
20
30
40
50
60
70
80
90
100
Photon Energy/eV
2.0
1.5
1.0
0.5
0.0
-0.5
13
14
15 16
17
18
19
20 21
22
23
(
c)
Photon Energy/eV
Fig. 4. Asymmetry parameters for S atoms plotted as a function of photon energy and compared with the results of Dill et al. [7]: (a) for the first S
2
þ
4
3
þ
2
3
þ
3
atom band, S ( S) S( P), (b) for the second S atom band, S ( D) S( P), and (c) for the third S atom band, S ( P) S( P).

L. Zuin et al. / Chemical Physics 298 (2004) 213–222
219
2
1
1
0
0
.0
.5
.0
.5
.0
3
0
+
4
3
S ( S)
S( P)
+
3
-
2
(
a) SH (X Σ )
SH(X Π)
20
10
0
-
0.5
1.0
-
2.0
1.5
1.0
0.5
0.0
13
14
15
16
17
18
+
1
2
(
b) SH (a ∆)
SH(X Π)
-
-
0.5
1.0
Photon Energy/eV
(
a)
+
2
3
S ( D)
S( P)
2
1
1
0
0
.0
.5
.0
.5
.0
6
5
4
3
2
1
0
0
0
0
0
0
+
1
+
2
(
c) SH (b Σ )
SH(X Π)
-
0.5
1.0
-
13
14
15
16
17
18 19
20
21
Photon Energy/eV
Fig. 5. Plots of the asymmetry parameter (b) as a function of photon
energy over the range hm ¼13:0–18.5 eV, for the first three photo-
electron bands of SH. (b) includes b values for the first band of H2S
measured at different photon energies in [34], denoted with the symbol
1
3.2
13.3
13.4
13.5
13.6
13.7
13.8
(*) in the figure.
(b)
Photon Energy/eV
þ
4
3
Fig. 6. (a) Relative cross-sections for the S ( S) S( P) band in the
photon energy region 13.205–18.494 eV. The maximum observed at
3.30 eV corresponds to excitation to unresolved [3s 3p (2P) nd] 3
P
2
3
1
with the results of the calculations [7] for the
þ4
þ
2
3
Rydberg states. (b) Relative cross-sections for the S ( D) S( P)
band in the photon energy region 13.205–13.412 eV. The resonances
present in the region 13.205–13.350 eV have been assigned to excita-
3
þ2
3
S ( S) S( P) and S ( D) S( P) ionization channels,
þ2
3
but are in poorer agreement for the S ( P) S( P)
ionization. However, for all three ionizations data points
are required to higher photon energy than the range
covered in this work (notably in the 25–50 eV region) for
a more detailed comparison between experiment and
theory. Fig. 3(b) also shows data points taken from [34]
for the first band of H2S. This shows reasonable agree-
ment with the S atom results of this work but again more
H2S data points are required for a more meaningful
comparison.
tions to [3s 3p (2P) nd] 3P Rydberg states which are part of a series
2
3
which converges to the third ionization limit at 13.400 eV.
In Figs. 6 and 7 the relative photoionization cross-
sections for the first three S atom bands are plotted as a
function of photon energy in the energy region 13.0–21.7
eV. In Fig. 6(b), the relative cross-section of the second
þ2
3
S atom band, the S ( D) S( P) ionization, is plotted
as function of the photon energy in the interval 13.20–
13.80 eV. As can be seen, this plot shows clear structure
in the region 13.205–13.350 eV. The observed maxima
Fig. 5 shows similar b-plots for the first three bands
of SH, with Fig. 5(b) again showing the available data
for the first band of H S from [34]. No previous mea-
surements or calculations of the b-parameter as a
function of photon energy have been made for SH.
However, as with the S-atom data, more data points are
required notably at higher photon energy. Also, it
should be borne in mind that, as stated earlier, the first
band of SH contains a small contribution from H2S and
this will contribute to Fig. 5(a).
2
3
can be assigned to excitation to P Rydberg states de-
2
3
rived from [3s 3p nd] configurations accessed from the
3s 3p4 3P ground state. These Rydberg states are parts
of a series which converges to the third ionization limit
at 13.400 eV [12,13].
2
To estimate the behaviour of the absolute photoion-
ization cross-section for the three ionic states of sulphur,

220
L. Zuin et al. / Chemical Physics 298 (2004) 213–222
+
4
3
+
1
2
3
2
1
0
0
0
40
SH (a ∆)
SH(X Π)
(
a) S ( S)
S( P)
35
30
25
20
0
0
4
3
2
1
+
2
3
(
b) S ( D)
S( P)
15
0
0
0
0
10
13.2
13.3
13.4
13.5
13.6
13.7
13.8
(
a)
Photon Energy/eV
+
3
-
2
57
(1) SH (X Σ )
SH(X Π)
1
1
1
4
2
0
8
6
4
2
0
+
2
3
42
(c) S ( P)
S( P)
2
8
4
1
35
+
1
2
(
2) SH (a ∆)
SH(X Π)
1
3
14
15
16
17
18
19
20
21
28
21
Photon Energy/eV
1
4
7
0
Fig. 7. Relative photoionization cross-sections for the first three bands
of S atoms in the photon energy region 13.0–18.5 eV. (b) includes r
values for the first band of H2S measured at different photon energies
in [38], denoted with the symbol (*) in the figure.
+
1
+
2
8
6
3
0
(
3) SH (b Σ )
SH(X Π)
the integrated area of each band was estimated at a
photon energy of 16.7 eV, an energy at which the total
photoionization cross-section has been calculated by
Yeh and Lindau [9] as 18.2 Mb, using Hartree–Fock–
Slater calculations. While for the second and third
1
3
14
15
16
17
18
19
20
21
þ2
3
(b)
Photon Energy/eV
bands, corresponding to the S ( D) S( P) and the
þ2
3
S ( P) S( P) ionizations, the integrated areas could be
easily measured, for the first S atom band, the
Fig. 8. Relative photon ionization cross-sections for SH plotted as a
function of photon energy (a) for the second band of SH in the photon
energy region 13.205–18.494 eV, and (b) for the first three bands of SH
in the photon energy region 13.0–18.5 eV. (b,2) includes r values for
the first band of H2S measured at different photon energies in [38],
denoted with the symbol (*) in the figure.
þ4
3
S ( S) S( P) ionization, that strongly overlaps with
the first band of SH, the result of the double-Gaussian
deconvolution shown in Fig. 2(b) was used. The con-
tribution of the first H2S band at 10.48 eV is included
with the first band of SH in the second Gaussian band
and it does not affect the calculated intensity of the first
S band, which corresponds to the first fitted band. As a
result, the intensity ratios of the first three bands of S
ꢁ
1
ized to the total cross-section of the (3p) ionization of
18.2 Mb, calculated by Yeh and Lindau [9].
Once this had been done, the experimental results
could be used to estimate the photoionization cross-
section at 13.8 eV, an energy at which the total photo-
ionization cross-section has been determined in an
experimental study of the absorption spectrum of S at-
oms. In this investigation, Tondello [10] determined the
photoionization cross-section over the energy region
investigated, 10.3–13.8 eV. In the photon energy region
0
atoms, were measured at an angle of 54°44 , as
(
6.2 ꢂ1.0):10:(5.7 ꢂ0.5) at 16.7 eV. The relatively high
experimental uncertainty of the first band relative in-
tensity is due to overlap of the first S atom band with the
first bands of SH and H2S, as stated earlier. In order to
estimate the absolute cross-sections for these ionizations
at 16.7 eV, the total areas of these bands were normal-

L. Zuin et al. / Chemical Physics 298 (2004) 213–222
221
1
2.8–13.8 eV, which overlaps with the photon energy
This initial study of S and SH has shown that PES
studies on short lived species of this type are possible
using synchrotron radiation, and angular distribution
and relative cross-section measurements can be made.
For SH, the results obtained are new and no calcula-
tions of r or b as a function of photon energy are
available with which the results obtained can be com-
pared. For S atoms no previous measurements of this
type have been made but, calculations are available of
r and b as a function of photon energy [6–8]; however,
the measurements need to be extended over a wider
photon energy range and should be recorded with a
smaller step-size to allow a more detailed comparison
to be made. For example, investigation of the structure
shown in Fig. 6(b) at higher resolution should allow
the Fano profiles of the observed bands to be investi-
gated. Also, it would be valuable to put the relative
cross-section plots obtained in this work for S atoms
on an absolute scale. This should be possible as the
photoionization cross-section of the first band of H2S
as a function of photon energy is known in the 10.0–
60.0 eV region from dipole (e, e + ion) coincidence
spectroscopy [38] and at some selected wavelengths
in the vacuum ultraviolet region from photoionization
measurements [39]. For example, the photoionization
cross-section of the first band of H2S has been mea-
sured at 16.7 and 21.22 eV as 17.65 and 9.30 Mb, re-
spectively [38]. Therefore, if spectra were recorded of
the reaction F + H2S under conditions which converted
region investigated in the present work, the photoioni-
zation cross-section decreases regularly from 90 to 79
Mb. At 13.8 eV, values for the photoionization cross-
sections of the first two bands of atomic S could be
obtained from the plots made in the present study,
whereas the photoionization cross-section for the third
band was obtained by extrapolation of the plot shown in
Fig. 7(c) to lower energy. This gave estimates of the
photoionization cross-section for the first three bands of
S atoms at 13.8 eV, which when summed gave a total
photoionization cross-section of 58 Mb, which can be
compared with the photoionization cross-section value
determined in TondelloÕs [10] work of 79 Mb. Given the
approximations involved, most notably that the total
photoionization cross-section in the present work has
been referenced to the photoionization cross-section
calculated at 16.7 eV by Hartree–Fock–Slater calcula-
tions [9], it is not surprising that only moderate agree-
ment is obtained. Nevertheless, this comparison
indicates there is a clear need for a direct measurement
of the photoionization cross-section of atomic sulphur
in the vacuum ultraviolet region, most notably above
the third ionization energy of 13.4 eV, i.e., at photon
energies which include all three ionic states arising from
ꢁ
1
the (3p) ionization.
In Fig. 6(a), the relative photoionization cross-section
of the first S atom band is plotted as a function of the
photon energy in the interval 13.205–18.494 eV. In this
figure the maximum at 13.30 eV corresponds to excitation
to the [3s 3p ( P) nd] P Rydberg states seen in Fig. 6(b),
recorded with a smaller photon energy increment. The
general behaviour of the photoionization cross-sections
for the first three ionizations of sulphur atoms are shown
in Fig. 7. In this figure it can be seen that except for the
resonance at 13.30 eV, already assigned to excitation to
all the H S to S atoms (with negligible subsequent re-
2
2
3
2
3
actions, e.g., SF production), and the spectrum of H S
2
2
was then recorded with the F /He discharge turned off
2
with all the other experimental conditions unchanged,
then the cross-section of the S bands can be determined
from the known cross-section of the first H2S band at
the photon energy used. Experiments of this type are
proposed.
2
3
2
3
[3s 3p ( P)nd] P Rydberg states, at least two other small
maxima are seen at 14.96 and 15.37 eV. These probably
1
4
4
3
3
3
correspond to excitation to [3s 3p ( P) np] D/ P/ S or
2
1
4
2
3
3
3
4
[
(
3s 3p ( P) np] D/ P/ S states accessed from the 3s 3p
3P) state. However, because of the step-size used to re-
4. Conclusions
cord these spectra (0.20 eV) and the limited number of
data points, assignment of these features will only be
achieved when spectra of the type shown in Figs. 6 and 7
are recorded over a wider photon energy range and with a
smaller step-size.
Figs. 8(a) and (b) show plots of relative photoioni-
zation cross-sections for the first three bands of SH.
They show similar overall behaviour to that observed
for the equivalent plots for S atoms (Fig. 7) and H2S
In this initial investigation on S and SH with syn-
chrotron radiation, the angular distribution parameters
and relative photoionization cross-section for the
þ4
3
þ2
3
þ2
3
S ( S) S( P), S ( D) S( P) and S ( P) S( P)
2
þ
3
ꢁ
1 2
þ
ionizations and the SH (X R , v ¼0) SH(X P,
0
0
þ
þ
00
v ¼0), SH (a D, v ¼0) SH(X P, v ¼0) and
þ
1
þ
þ
2
00
SH (b R , v ¼0) SH(X P, v ¼0) ionizations have
been evaluated as a function of photon energy from
threshold to 21.64 eV, using radiation from the Elettra
synchrotron as the photon source. The results obtained
have been compared, where available, with previous
experiments and calculations. Further experiments are
proposed to extend these measurements to higher pho-
ton energies.
[
34], but, as with the S atom plots, need to be recorded
over a wider photon energy range with a smaller step-
size. Figs. 7(b) and 8(b,2) also include the experimental
photoionization cross-section for the first band of H2S
as a function of photon energy, taken from [38], for
comparison.

2
22
L. Zuin et al. / Chemical Physics 298 (2004) 213–222
[19] P.E. Cade, W.M. Huo, J. Chem. Phys. 47 (1967) 649.
Acknowledgements
[
20] M. Horani, J. Rostas, H. Lefebvre-Brion, Can. J. Phys. 45 (1967)
319.
21] J.B. Milan, W.J. Buma, C.A. de Lange, J. Chem. Phys. 105 (1996)
688.
3
The authors are grateful to the Reactive Intermedi-
ates RTN EC Network, EPSRC and the Leverhulme
Trust for support of this work. The authors also ac-
knowledge support from Dr. N. Zema and Dr. S. Tur-
chini and the staff of the POLAR beamline at Elettra.
[
6
[22] C.W. Hsu, D.P. Baldwin, C.L. Liao, C.Y. Ng, J. Chem. Phys. 100
(1994) 8047.
[
23] J.B. Milan, W.J. Buma, C.A. deLange, J. Chem. Phys. 104 (1996)
21.
5
[24] J.M. Dyke, S.D. Gamblin, A. Morris, T.G. Wright, A.E.
Wright, J.B. West, J. Electron Spectrosc. Relat. Phenom. 97
References
(1998) 5.
25] J.M. Dyke, S.D. Gamblin, A. Morris, T.G. Wright, A.E. Wright,
J.B. West, J. Phys. B 32 (1999) 2763.
[
[
1] L.J. Beeching, A.A. Dias, J.M. Dyke, A. Morris, S. Stranges, J.B.
West, N. Zema, L. Zuin, Mol. Phys. 101 (2003) 575.
[26] Available from <http://kinetics.nist.gov/index.php>.
[27] M.C.R. Cockett, J.M. Dyke, H. Zamanpour, in: C.Y. Ng (Ed.),
Vacuum Ultraviolet Photoionization and Photodissociation of
Molecules and Clusters, World Scientific, 1991 (Chapter 2).
[28] J.B. West, G.V. Marr, Atom. Data Nucl. Data Tables 18 (1976)
497.
[
2] (a) G.G. Lombardi, B.L. Cardon, R.L. Kurucz, Astrophys. J. 248
(
(
1981) 1208;
b) J.E. Vernazza, E.H. Averett, R. Loeser, Astrophys. J. Suppl.
45 (1981) 635.
[
[
[
3] M.L. Meeks, M.A. Gordon, M.M. Litvak, Science 163 (1969) 173.
4] K. Sinha, Proc. Astron. Soc. Aust. 9 (1991) 32.
5] S.J. Dunlavey, J.M. Dyke, N.K. Fayad, N. Jonathan, A. Morris,
Mol. Phys. 38 (1979) 729;
[29] (a) A. Derossi, F. Lama, M. Piacentini, T. Prosperi, N. Zema,
Rev. Sci. Instrum. 66 (1995) 1718;
(b) D. Desidero et al., Synchrotron Radiat. News 12 (1999)
34.
Mol. Phys. 44 (1981) 265.
[
[
[
6] M.J. Conneely, K. Smith, L. Lipsky, J. Phys. B 3 (1970) 493.
7] D. Dill, A.F. Starace, S.T. Manson, Phys. Rev. A 11 (1975) 1596.
8] S.T. Manson, A. Msezane, A.F. Starace, S. Shahabi, Phys. Rev. A
[30] A.B. Cornford, D.C. Frost, C.A. McDowell, J.L. Ragle, I.A.
Stenhouse, J. Chem. Phys. 54 (1971) 2651.
[31] P.M. Guyon, R. Spohr, W.A. Chupka, J. Berkowitz, J. Chem.
Phys. 65 (1976) 1650.
20 (1979) 1005.
[
10] G. Tondello, Astrophys. J. 172 (1972) 771.
9] J.J. Yeh, I. Lindau, Atom. Nucl. Data Tables 32 (1985) 1.
[32] D.M. deLeeuw, R. Mooyman, C.A. deLange, Chem. Phys. 34
(1978) 287.
[
[
[
11] V.N. Sarma, Y.N. Joshi, Physica 123C (1984) 349.
12] Y.N. Joshi, M. Mazzoni, A. Nencioni, W.H. Parkinson, A. Cantu,
J. Phys. B 20 (1987) 1203.
[33] C.E. Moore, Atomic Energy Levels, vol. 1, National Bureau of
Standards 467 US Government Printing Office, Washington, DC,
1949.
[
[
[
[
13] S.T. Gibson, J.P. Green, B. Ruscic, J. Berkowitz, J. Phys. B 19
[34] P. Baltzer, L. Karlsson, M. Lundqvist, B. Wannberg, D.M.P.
Holland, M.A. MacDonald, Chem. Phys. 195 (1995) 403.
[35] M.Y. Adam, P. Morin, C. Cauletti, M.N. Piancastelli, J. Electron
Spectrosc. Relat. Phenom. 36 (1985) 377.
(1986) 2825.
14] G.W. vanLoon, S.J. Duffy, Environmental Chemistry: A Global
Perspective, University Press, Oxford, 2000.
15] M.N.R. Ashfold, B. Tutcher, C.M. Western, Mol. Phys. 66 (1989)
[36] K. Watanabe, A.S. Jursa, J. Chem, Physics 41 (1964) 1650.
[37] (a) J.W. Cooper, R.N. Zare, Lect. Theor. Phys. 11C (1969)
323;
981.
16] G. Herzberg, K.P. Huber, In: Molecular Spectra and Molecular
Structure, vol. 4, D. van Nostrand Co. Inc., 1966.
17] B.A. Morrow, Can. J. Phys. 44 (1966) 2447.
(b) J.W. Cooper, Phys. Rev. 128 (1962) 681.
[
[
[38] R. Feng, G. Cooper, C.E. Brion, Chem. Phys. 249 (1999) 223.
[39] T.J. Xia, T.S. Chien, C.Y.R. Wu, D.L. Judge, J. Quant. Spectrosc.
Radiat. Transfer 45 (1991) 77.
18] J. Brzozowski, N. Elander, P. Erman, M. Lyyra, Phys. Scr. 10
(1974) 241.