Photoionization of the primary photoproducts of A(2∏)-excited CIO

Article abstract of DOI:10.1063/1.1516734

Photoionization study of the primary photofragments of chlorine monoxide (ClO) is presented. The primary photolysis products and the non-photolyzed ClO are probed by monochromatic, time-correlated vacuum-ultraviolet radiation from a laser produced plasma source. The results are compared to additional photolysis work on molecular chlorine.

Full text of DOI:10.1063/1.1516734

Photoionization of the primary photoproducts of A( 2 ∏ ) -excited ClO
R. Flesch, J. Plenge, S. Kühl, M. Klusmann, and E. Rühl
Citation: The Journal of Chemical Physics 117, 9663 (2002); doi: 10.1063/1.1516734
View online: http://dx.doi.org/10.1063/1.1516734
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/117/21?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Isotope effects and spectroscopic assignments in the non-dissociative photoionization spectrum of N2
J. Chem. Phys. 140, 194303 (2014); 10.1063/1.4873717
Ab initio chemical kinetics for reactions of ClO with Cl2O2 isomers
J. Chem. Phys. 134, 054307 (2011); 10.1063/1.3541353
Inner-valence photoionization of O ( D 1 ) : Experimental evidence for the 2 s 2 2 p 4 ( D 1 ) → 2 s 1 2 p 5 ( P 1 )
transition
J. Chem. Phys. 128, 074307 (2008); 10.1063/1.2830232
The ultraviolet photodissociation of jet-cooled ClO and BrO radicals
J. Chem. Phys. 116, 4176 (2002); 10.1063/1.1448282
Photodissociation of CH stretch overtone excited CH 3 Cl and CHD 2 Cl (v CH =5): Cl spin–orbit branching and
atomic fragment yields
J. Chem. Phys. 109, 7810 (1998); 10.1063/1.477427
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

JOURNAL OF CHEMICAL PHYSICS
VOLUME 117, NUMBER 21
1 DECEMBER 2002
2
Photoionization of the primary photoproducts of A„ ͟…-excited ClO
a)
¨
¨
R. Flesch, J. Plenge, S. Kuhl, M. Klusmann, and E. Ruhl
¨
¨
¨
Fachbereich Physik, Universitat Osnabruck, Barbarastrasse 7, D-49069 Osnabruck, Germany
͑Received 12 August 2002; accepted 30 August 2002͒
Photoionization of the primary photofragments of chlorine monoxide ͑ClO͒ is reported. ClO is
photolyzed in the X(2
͟
)→A(2
͟
)-regime, yielding Cl(2P) and O(3P,1D). The primary photolysis
products, as well as the not photolyzed ClO, are subsequently probed by monochromatic,
time-correlated vacuum-ultraviolet radiation from a laser produced plasma source. Autoionization is
used for state-specific detection of the atomic photolysis products. The formation of O(3P) is
exclusively observed above Ϸ264 nm. The threshold of O(1D) from A(2
͟
3/2)-excited ClO is found
at 263.71Ϯ0.01 nm. The shape of the O(1D) yield near this threshold is discussed in terms of the
rotational energy distribution and a rotational barrier of A(2
3/2)-excited ClO. Direct ͑nonresonant͒
one-photon-ionization is used to establish the absolute photoionization cross sections of
͟
ClO(X(2 )), Cl(2P), and O(1D) near 15 eV. Additional experiments on the UV-photolysis of Cl2,
͟
yielding Cl(2P), are consistent with the results on ClO. The present work is compared to earlier
experimental and theoretical studies. © 2002 American Institute of Physics.
͓DOI: 10.1063/1.1516734͔
I. INTRODUCTION
͑ii͒ Primary quantum yields, partial photolysis quantum
yields, and branching ratios of competing photolysis pro-
cesses are accurately determined from single-photon ioniza-
tion, if the absolute photoionization cross sections of the spe-
cies that are ionized are known.16
Molecular photofragmentation plays an important role in
the field of atmospheric chemistry and physics.1 The atmo-
spheric trace gas budget is strongly influenced by the release
of reactive species that come from photolysis processes,
where ultraviolet and visible radiation are most efficient.
Photodissociation into free atoms, such as oxygen or chlo-
rine, is of particular interest to atmospheric research, since
these species are highly reactive and initiate subsequent
chemical reactions in the atmosphere, leading, e.g., to polar
ozone destruction in the stratosphere.2–5
The dissociation of reactive, electronically excited states
of atmospheric molecules and radicals has been investigated
by a variety of experimental approaches in the past. This
included resonance enhanced multiphoton ionization,6–8
fluorescence spectroscopy,9,10 and photofragment transla-
tional energy spectroscopy.11,12 Alternatively, photoioniza-
tion pump–probe approaches have been developed recently,
where the primary photolysis products are probed by mono-
chromatic vacuum ultraviolet ͑VUV͒ radiation.13 This leads
to single-photon ionization so that the charged products can
be probed by mass spectrometry. Single-photon ionization
has the following advantages compared to multiphoton ion-
ization approaches.14,15
͑iii͒ Absolute photoionization cross sections (␴I) can be
determined in cases where the stoichiometry of the photoly-
sis processes is known. For example, we have established
previously the photoionization cross sections of chlorine di-
oxide ͑OClO͒ and chlorine monoxide ͑ClO͒.17
We report in this study recent results on the photolysis of
chlorine monoxide ͑ClO͒, which is also a suitable photolytic
precursor of O(3P), O(1D), and Cl(2P), respectively. ClO
is known to occur in the stratosphere, where it plays a key
role in the context of polar ozone depletion.1,2,18 The photo-
absorption cross section of ClO in the ultraviolet shows an
extended rovibronic progression between 260 and 320 nm,
corresponding to the X(2
͟
)→A(2
͟
)-transition.19 All dis-
result of fast
crete features are broadened as
a
predissociation,20 which has also been the subject of recent
theoretical work.21,22 The highest assigned vibrational level
in the progression is ϭ25,20 which is followed by the di-
v
Ј
rect dissociation limit of the A(2
͟) state near ␭ϭ263 nm.
The ground state atomic products are formed in the bound
region at ␭у263 nm:
͑i͒ Ionic fragmentation is controlled or entirely sup-
pressed by tuning the VUV radiation into the threshold re-
gime, which is commonly termed ‘‘soft ionization’’. Ionic
fragmentation is more difficult to control in the case of mul-
tiphoton laser ionization where photoabsorption in ionization
continua may lead to extensive ionic fragmentation, because
of the nonlinear optical processes that may be involved.
2
2
2
3
͑ ͑
͒͒ϩh␯→ClO A
͒͒→Cl P͒ϩO P͒.
͑
͑
ClO X
͑ ͑
͟
͟
͑1͒
The
dissociation
limit
of
the
X(2
͟
)
3/2
→A(2
͟
3/2)-progression has been derived from optical spec-
troscopy, yielding ␯ϭ38 052 cm (␭ϭ262.798 nm͒.20 The
formation of excited O(1D) has been shown to be the result
of direct photodissociation:23
Ϫ1
˜
a͒
¨
¨
¨
Current address: institut fur Physikalische Chemie, Universitat Wurzburg,
2
2
2
1
ClO X
͑ ͑
͟
͟
͒͒ →Cl P͒ϩO D͒,
͓
͑ ͑
͒͒ϩh␯→ ClO A
͔
͑
͑
¨
Am Hubland, D-97074 Wurzburg, Germany. Electronic mail:
eruehl@phys.chemie.uni-wuerzburg.de
͑2͒
0021-9606/2002/117(21)/9663/8/$19.00
9663
© 2002 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

9664
J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Flesch et al.
where the square brackets indicate that the continuum of the
A(2
) state is excited at ␭Ͻ263 nm. This yields the products
Cl(2P)ϩO(1D), corresponding to the asymptote of the
A(2
) state. The photoprocesses 1 and 2 mutually exclude
͟
͟
each other, so that electronically excited ClO is either a
source of Cl(2P)ϩO(3P) or Cl(2P)ϩO(1D). However, re-
cent work indicates that the photolysis channel yielding
Cl(2P)ϩO(3P) also occurs with weak intensity at ␭ϭ248
nm.11
The proper description of the photoionization cross sec-
tion of atomic chlorine presents a challenge for theory.24 The
first ionization energy of atomic chlorine (2P3/2) is 12.9676
eV.25 As a consequence of the open-shell structure of atomic
chlorine, 3p photoionization leads to the electron configura-
tion 3s23p4 which gives rise to three terms, i.e.. P2,1,0 ͑0,
3
0.086 293, 0.123 547 eV͒, 1D2 ͑1.444 860 eV͒, and 1S0
FIG. 1. Photoionization mass spectra of an OClO/Cl2 mixture diluted in He:
͑a͒ Photoionization of the stable reactants at Eϭ13.10 eV; ͑b͒ Formation of
ClO from the gas mixture by using a microwave discharge ͑see the text for
further details͒.
͑3.456 435 eV͒.26 The experimental photoion yield of Cl(2P)
3
has been reported earlier in the regime between the P and
1S thresholds.27 It is characterized by a variety of autoioniz-
ing resonances. They are assigned to nd and ns series con-
verging to the D and S state of Clϩ, respectively.28 A va-
riety of experimental and theoretical studies on
photoionization of Cl(2P) has been published before27,29–31
͑cf. Sec. III C͒.
1
1
mass spectrometer serves for cation separation and detection.
It is mounted behind the exit slit of the vacuum monochro-
mator.
Chlorine monoxide is prepared by the reaction of atomic
chlorine with chlorine dioxide ͑OClO͒ in a fast flow tube at
room temperature, according to the preparation procedure
outlined in Ref. 34 ͑cf. Fig. 1͒:
One aim of the present study is to determine the O(1D)
yield close to the onset of the continuum that is associated
with electronically excited ClO(A(2 )), where O(1D) is se-
͟
lectively probed by single-photon ionization, similar to pre-
vious work.16,32 The absolute photoionization cross sections
of ClO and its photofragments, i.e., Cl(2P) and O(1D) are
determined using the same experimental approach. The
present results are compared to additional photolysis work
on molecular chlorine, which gives another independent ap-
proach to determine the absolute photoionization cross sec-
tion of Cl(2P). The results are compared to previous theo-
retical work, where one-photon ionization cross sections are
also of interest to the physics of the ionosphere and free
space.33
OClOϩCl→2ClO.
͑3͒
Atomic chlorine is produced in a microwave discharge,
where a 5% mixture of molecular chlorine in helium is used.
OClO is prepared by the reaction of molecular chlorine with
sodium chlorite.35 The yellow gas OClO is dried over phos-
phorus pentoxide and is used without further purification for
the experiments. The other compounds that are used in this
work are of commercial quality (Cl2 :98% ͑Aldrich͒,
NaClO2 :80% ͑Aldrich͒, He: 99.99% ͑Messer Griesheim͒͒.
They are used without further purification.
Figure 1 shows photoionization TOF mass spectra of an
OClO/Cl2/He mixture, where the photon energy is set to
13.10 eV. Figure 1͑a͒ displays a mass spectrum of the gas
mixture without the production of atomic chlorine, whereas
the mass spectrum shown in Fig. 1͑b͒ corresponds to condi-
tions where atomic chlorine is efficiently produced in the
microwave discharge from the same gas mixture. Both mass
spectra are dominated by intense signals that are due to
photoionization of molecular chlorine (IE͑Cl2)ϭ11.48
eV36͒. Additional mass signals in Fig. 1͑a͒ are due to photo-
ionization of OClO ͑IE͑OClO͒ϭ10.33 eV37͒. The mass sig-
nal of ClOϩ is of negligible intensity in Fig. 1͑a͒, since the
threshold for ClOϩ formation from OClO is 13.40Ϯ0.05
eV.37 The spurious ClOϩ mass signal comes most likely from
second order light of the plasma source, which contributes
Ͻ5% to the VUV photon flux. Figure 1͑b͒ shows evidence
for the efficient formation of ClO in the flow system ͓cf. Eq.
͑3͔͒, since a considerable ClOϩ-signal is formed as a result
of photoionization above first ionization energy of ClO
͓IE͑ClO͒ϭ10.85Ϯ0.05 eV38͔. Consistently, the OClOϩ sig-
nal has vanished, since this reactant is entirely consumed by
II. EXPERIMENT
A detailed description of the experimental setup has been
published elsewhere.32 Briefly, a tunable, excimer-pumped,
pulsed dye-laser ͑Lambda Physik: Compex 201 and Scan-
mate͒ equipped with a frequency doubling BBO͑I͒ crystal is
used for UV excitation and photolysis of gas phase mol-
ecules under collision-free conditions.
A pulsed, tunable laser-produced plasma ͑LPP͒ radiation
source provides intense, monochromatic VUV radiation in
the energy regime between 8 eVрEр25 eV, corresponding
to 50р␭р155 nm. The radiation consists essentially of a
broad background continuum, where tungsten is used as a
target. The divergent radiation is focused onto the entrance
slit of a 1 m vacuum monochromator ͑McPherson, Nova
225͒ equipped with a 1200 lines/mm holographically ruled
gold-coated grating ͑Jobin Yvon͒. The bandwidth of the LPP
radiation is estimated to be of the order of 1.5 nm. LPP
radiation is used to photoionize the photolysis products as
well as the molecular precursors. A time-of-flight ͑TOF͒
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Photoionization of photoproducts of ClO
9665
the ClO synthesis. The present results indicate, that the
sample which is used for subsequent photolysis studies con-
sists of a Ϸ1:1 mixture of ClO and Cl2 in a gas flow that
contains Ϸ95% helium.
The experiments are carried out as follows: Photoioniza-
tion mass spectra are obtained from the VUV-radiation
source with and without primary photolysis of the neutral gas
phase. The photon energy is tuned in order to obtain suffi-
cient selectivity for the final photoionization step. A typical
set of mass spectra consists of ͑a͒ a photoionization mass
spectrum without primary photolysis; ͑b͒ a pump–probe
photoionization mass spectrum with primary photolysis and
subsequent, time-correlated photoionization; ͑c͒ a photoion-
ization mass spectrum obtained exclusively from the pho-
tolysis laser, so that contributions from multiphoton ioniza-
tion become evident. Data reduction is carried out by
subtracting the mass spectra ͑a͒ and ͑c͒ from ͑b͒. The result-
ing mass spectra are termed photolysis mass spectra in the
following, since they contain exclusively the result of
photofragmentation of the parent molecule. As a result, par-
ent molecules occur with negative signal intensity in pho-
tolysis mass spectra because they are consumed by UV pho-
tolysis. It is noted that the gas phase synthesis of ClO in the
flow tube did not show any aging effects of the halocarbon
wax-coating.
FIG. 2. Photolysis mass spectra of ClO recorded at ␭ϭ273.2 nm: ͑a͒ photo-
ionization at Eϭ14.10 eV; ͑b͒ photoionization at Eϭ14.60 eV. See the text
for further details.
tion of the atomic photolysis products. The relative cation
intensities depend strongly on the photon energy that is used
for photoionization. Figure 2͑a͒ has been recorded at 14.10
eV, whereas Fig. 2͑b͒ corresponds to photoionization at
14.60 eV. A comparison between both photolysis mass spec-
tra indicates that the Oϩ signal is considerably more intense
in Fig. 2͑a͒. This result is interpreted as follows: The ioniza-
tion energy of O(3P) is IE͑O͒ϭ13.61806 eV.43 The autoion-
Additional experiments are carried out by photolyzing
molecular chlorine at ␭ϭ308 nm by using a XeCl excimer
laser.
izing
multiplet
transition
1s22s22p4
(3P)
→1s22s22p33s (3P) occurs near Eϭ14.10 eV,41 so that
Љ
O(3P) can be efficiently probed by autoionization yielding
an intense Oϩ signal ͓cf. Fig. 2͑a͔͒. In contrast, there is no
autoionization resonance near 14.60 eV. Hence, photoioniza-
tion can occur only as a result of direct ͑nonresonant͒ photo-
ionization, which yields a weak Oϩ signal ͓cf. Fig. 2͑b͔͒.
This indicates that the direct photoionization cross section is
considerably weaker than that at autoionization resonances.44
The Oϩ signal, which is obtained from photoionization
at Eϭ14.10 eV, is selected while scanning the photolysis
wavelength. This yields the relative cross section of O(3P)
III. RESULTS AND DISCUSSION
A. State-selective detection of the atomic photolysis
products from ClO
The tunability of the VUV radiation allows us to mea-
sure photoion yields of mass-selected cations.32 Autoioniza-
tion features are used to obtain further information on the
quantum state of atomic photolysis products. Specifically, we
have recently shown that photoionization of photolytically
formed O(1D) leads to a variety of autoionization reso-
nances which are due to transitions into autoionizing singlet
Rydberg states.39 The Oϩ photoion yield from O(1D) has
been compared recently to the Oϩ photoion yield obtained
from O(3P),40 where the O(3P) yield has been reported
earlier.41 It is found that autoionizing resonances of both spe-
cies are different from each other, so that atomic oxygen can
be probed state-selectively by autoionization.
formation from A(2
͟)-excited ClO, as shown in Fig. 3. The
Figure 2 shows photolysis mass spectra of ClO that are
measured under pump–probe conditions, where the ClO is
photolyzed at ␭ϭ273.2 nm. The photolysis wavelength cor-
responds
to
the
X(2
͟
(
ϭ0))→A(2
͟
(
ϭ13))
v
Ј
v
Љ
transition.42 This wavelength is well above the Cl(2P)
ϩO(1D) formation threshold so that only the products
Cl(2P)ϩO(3P) are expected to occur. Subsequently, the
mixture containing ClO as well as the atomic photolysis
products is photoionized at various VUV-photon energies.
Figure 2 clearly shows that the ClOϩ signal becomes
negative, which indicates that ClO is consumed by UV pho-
tolysis. Consequently, there is a gain of the atomic cation
intensities Oϩ and Clϩ, which are formed by photoioniza-
FIG. 3. ͑a͒ Photoabsorption cross section of ClO ͑data taken from Ref. 42͒;
͑b͒ O(3P) yield from primary ClO photolysis and subsequent photoioniza-
tion at Eϭ14.10 eV.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

9666
J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Flesch et al.
FIG. 4. Photolysis mass spectra of ClO that evidence the formation of
O(1D) at ␭ϭ250 nm. Experimental conditions: ͑a͒ photoionization at E
ϭ13.45 eV. The strong Oϩ signal is due to autoionization of O(1D). ͑b͒
Photoionization at Eϭ13.75 eV.
FIG. 5. Primary quantum yield of O(1D) formation from
ClO͑A(2 3/2))␸3/2(O(1D)). The closed circles correspond to the experi-
͟
mental data. The solid line ͑a͒ is a calculated threshold of O(1D) from
results indicate that the Oϩ yield follows between 270 and
274 nm the photoabsorption cross section of ClO ͑cf. Ref.
42͒. Figure 3 also indicates that the photolytically formed
atomic oxygen (O(3P)) comes exclusively from the primary
photolysis of ClO via Eq. ͑1͒.
Љ
thermal sample at Tϭ300 K, where all rotational levels R of ground state
ClO͑A(2
͟3/2)) contribute. Line ͑b͒ represents a sigmoidal fit to the experi-
mental data in order to guide the eye. The absorption cross section is in-
cluded for a comparison at the top, where the data are taken from Ref. 45.
See the text for further details.
Figure 4 shows photolysis mass spectra of ClO that are
obtained from a different set of excitation conditions, where
the photolysis wavelength is set to ␭ϭ250 nm. This excita-
tion wavelength is below the dissociation limit of the X
→A progression which is known to vanish at ␭р263 nm.20
The bottom trace ͓Fig. 4͑a͔͒ is obtained from photoionization
at Eϭ13.45 eV. This energy corresponds to autoionization of
The result is shown in Fig. 5, where the Oϩ yield is
normalized to the flux of the photolysis laser and the VUV
radiation source. The experimental data are represented by
closed circles. A sigmoidal fit is used to indicate the steep
onset of the Oϩ yield ͓see curve ͑b͒ in Fig. 5͔. This onset is
compared to the absorption cross section, which is shown at
the top of Fig. 5.45 The density of vibrational states near the
dissociation limit is high. The spacings between the states is
O(1D) via the multiplet transitions 2p4(1D)→2p3(2D)3d
Ј
where the corresponding singlet Rydberg states are known to
be efficiently autoionized by the Oϩ(4S) continuum, yield-
ing an intense Oϩ signal.39 Moreover, the photoionization
cross section of atomic chlorine (Cl(2P)) is also enhanced
by autoionization at this photon energy ͑cf. Ref. 27͒, which
explains the intense Clϩ signal in Fig. 4͑a͒. The photolysis
mass spectrum shown in Fig. 4͑b͒ is recorded at the same
photolysis wavelength, but the VUV-photon energy is
slightly increased to Eϭ13.75 eV, where no O(1D) autoion-
ization occurs according to Ref. 39. Moreover, no evidence
for O(3P) formation is found under these photolysis condi-
tions. As a result, no Oϩ signal is observed. The Clϩ signal
is decreased by a factor of ϳ3 compared to Fig. 4͑a͒. This is
in excellent agreement with the shape of the photoionization
cross section of Cl(2P) ͑cf. Ref. 27͒.
of the order of several tens of wave numbers: The origins ␯
0
of discrete vibrational levels ( ϭ22, 23, 24, and 25͒ are
v
Ј
˜
found at ␯ϭ37 917.13, 37 963.37, 37 997.36, and 38 021.06
cmϪ1, respectively ͑cf. Fig. 5͒, where ϭ25 is the highest
v
Ј
vibrational level that has been resolved in optical spectra.20
These levels are evidently located above the threshold re-
gime of O(1D) formation from thermal ClO ͑see Fig. 5͒. It
was also shown before that another five vibrational levels
Ϫ1
˜
occur between ϭ25 (␯ϭ38 021.06 cm ͒ and the disso-
v
Ј
ciation threshold (␯ϭ38 052 cm ͒.20
Ϫ1
˜
The experimental data indicate that the onset of O(1D)
Ϫ1
˜
formation occurs at ␯ϭ37 920Ϯ2 cm (␭ϭ263.71Ϯ0.01
nm, see Fig. 5͒, corresponding to 10% of the maximum pri-
mary quantum yield ␸3/2(O(1D)). It is noted that the onset
of O(1D) formation shown in Fig. 5 is only due to the direct
1
2
B. Threshold of O„ D… formation from A„
͟
…-excited
ClO
The present approach is suitable to detect the threshold
of O(1D) formation that is known to occur near 263 nm.7,20
This is accomplished by measuring the yield of O(1D) for-
dissociation of the ⍀ϭ3/2 component of A(2
͟⍀)-excited
ClO. No attempt has been made to detect the onset of the
dissociation continuum that comes from excitation of the
⍀ϭ1/2 spin–orbit component. It is well-known from theory
mation from A(2
͟)-excited ClO in the 261 nmр␭р265 nm
regime, where the Oϩ signal is detected by tuning the pho-
tolysis laser, while keeping the VUV-photon energy at E
ϭ12.41 eV. This VUV photon energy probes exclusively
O(1D) via autoionization.39
and experiments that both A(2
͟⍀) spin–orbit states yield the
atomic fragments in the same fine structure states, i.e.,
Cl(2P3/2)ϩO(1D).11,23 The onset of direct dissociation from
the ⍀ϭ1/2 system is expected to be redshifted by the spin–
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Photoionization of photoproducts of ClO
9667
petes with dissociation through this barrier into Cl(2P)
ϩO(1D). This is not the case for low rotational levels where
the O(1D) yield increases steeply, so that no appreciable
rotational barrier exists for these levels, explaining qualita-
tively the steep onset of the O(1D) yield. A quantitative de-
scription of the results appears to be difficult since the shape
of the excited state potential near the dissociation threshold
is not well-known.
orbit splitting of the electronic ground state, i.e., by 318
cmϪ1 ͑cf. Ref. 46͒.
The dissociation limit of A(2
͟3/2)-excited ClO has been
derived earlier from the analysis of the photoabsorption cross
20
section, corresponding to ␯ϭ38 052 cmϪ1 (␭ϭ262.79 nm͒
˜
and ␯ϭ38 026Ϯ7 cm (␭ϭ262.90Ϯ0.05 nm͒.47 We note
Ϫ1
˜
that a threshold of O(1D)-formation at ␭ϭ263.4Ϯ0.2 nm
Ϫ1
˜
(␯ϭ37 965Ϯ29 cm ͒ was observed earlier in multiphoton
The occurrence of rotational barriers in molecular pho-
todissociation is in agreement with previous work, as ob-
served for the photolysis of O2 in the regime of the
Schumann-Runge bands (X(3͚uϪ)→B(3͚Ϫg )).52 Direct dis-
ionization work.7 This value corresponds to the half-
˜
maximum wave number ␯ of the experimental threshold
1/2
͑see Fig. 5͒. It is also noted that the present results have a
significantly improved signal-to-noise ratio, so that O(1D)
intensity is observed at longer photolysis wavelengths com-
pared to previous work.7
Ϫ
3
sociation of B ͚g -excited O2 via the Schumann–Runge
continuum results in the atomic products O(1D)ϩO(3P),
whereas predissociation of the Schumann–Runge bands
leads to the ground state atoms O(3P)ϩO(3P). The thresh-
old regime of O(1D)ϩO(3P) formation from highly excited
Schumann–Runge bands is characterized by a step-like
structure.52 It is due to rotational thresholds for the transition
from the X(3͚Ϫu )-ground state to energies above the
The primary quantum yield of O(1D) formation from
ClO(A(2 3/2)) ␸3/2(O(1D)) is expected to be zero below the
͟
threshold, corresponding to the constant Oϩ yield at ␭
Ͼ263.71 nm. It increases to a value close to unity above the
O(1D)-threshold, according to photofragment translational
energy spectroscopy,8,11 where the competing channel, lead-
ing to O(3P)ϩCl(2P3/2) formation, is found to remain weak.
This is in agreement with the present results. These findings
Ϫ
3
B
͚g -dissociation limit. Previous work on the photolysis of
molecular iodine (I2) at the B (3
͟
) dissociation threshold
ϩ
o
u
also indicates, that the rotational distribution at Tϭ300 K
gives rise to a rotational barrier. This leads to a sigmoidal
threshold shape, which is broadened by collisions.53 More
recent photolysis experiments on jet-cooled, rotationally cold
justify the absolute scale for
ClO(A(2
3/2)), which is included on the left-hand side of
Fig. 5. We estimate the O(1D) yield below the A(2
␸
3/2(O(1D)) from
͟
͟
)
3/2
threshold to be of the order of 0.2, which is a result of the
photolysis of the ⍀ϭ1/2 spin–orbit component.
hydrogen indicate that the threshold region of B (1͚ϩ) pho-
Ј
u
todissociation is quite narrow and covers a range of Ͻ2
The experimental onset of O(1D) formation is compared
to the calculated threshold behavior, which is obtained from
using the BCONT program48 ͓see Fig. 5, curve ͑a͔͒. The cal-
culation is based on the following considerations: The ClO
sample is prepared at room temperature (Tϭ300 K͒ in the
electronic ground state, as described in the Sec. II.34,49 The
potential of the ground and excited states are taken from Ref.
47, the transition moments are obtained from Ref. 50, and
the maximum of the curve has been adjusted to the experi-
mental data. The ground state rotational levels are thermally
cmϪ1 54
which also reflects the importance of rotational ex-
,
citation on the shape of photolysis thresholds.
Finally, we conclude, that the threshold of O(1D) forma-
tion is strongly affected by a rotational barrier that prevents
highly excited rotational states to populate the photolysis
channel ͑2͒.
C. Determination of absolute photoionization
cross sections
populated and the dissociation threshold of the A(2
therefore lowered by thermal excitation in the X(2
͟
) state is
͟
) ground
The relative cation intensity ISϩ of a species S in the
mass spectrum is proportional to the density Ns /V in the
ionization volume V that is detected by the mass spectrom-
state, leading to different dissociation thresholds from the
thermally excited rotational states. The energy positions of
the dissociation thresholds are also included in Fig. 5, where
the data have been taken from Refs. 47 and 51.
I
eter and the photoionization cross section ␴S :
NS
Curve ͑a͒ clearly shows that the threshold should tail to
considerably longer wavelengths than is observed in the ex-
periment. A thermal ClO sample is expected to have suffi-
cient internal thermal energy so that photolysis into O(1D)
ϩCl(2P3/2) can occur below the ϭ0, J ϭ3/2 threshold.
I
ISϩϭk •
•␴S ,
͑4͒
Ј
V
where k includes the light intensity and the detection sensi-
Ј
tivity of the photoionized species at the detector. It is as-
sumed in the following that the detection sensitivity is con-
stant for all species under investigation. Considering that a
molecule AB is photolyzed into the neutral fragments A and
B which are subsequently photoionized into Aϩ and Bϩ, the
v
Љ
Љ
For example, Ϸ30% of the thermal sample should have suf-
ficient rotational energy to be photolyzed via the A(2
͟
)
3/2
Ϫ1
˜
state at ␯ϭ37 800 cm ͑␭ϭ264.55 nm͒, where evidence for
O(1D) formation is just obtained from low and almost con-
stant Oϩ intensity. Thus, curve ͑a͒ shown in Fig. 5 appears to
have little resemblance to the experimental threshold behav-
ior of O(1D) formation.
I
A
I
B
photoionization cross section ␴ is related to ␴ as follows:
IBϩ•NA
IAϩ•NB
I
␴IBϭ␴A•
.
͑5͒
The present experiments indicate that there is likely evi-
dence for a rotational barrier for higher lying rotational lev-
els, so that predissociation yielding Cl(2P)ϩO(3P) com-
The particle density of the atomic products is unequivocally
determined by the stoichiometry of the photolysis process, as
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

9668
J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Flesch et al.
shown in Eqs. ͑1͒ and ͑2͒. Specifically, photolysis of ClO
results in the following budget: Ϫ⌬NClOϭNOϭNCl . Then,
Eq. ͑5͒ reads
It is noted that there are no autoionization resonances for
atomic oxygen (O(3P)) near Eϭ14.60 eV. The nearest auto-
ionization resonances are observed at Eϭ14.12 eV and E
ϭ15.15 eV, respectively.41 Therefore, no corrections of
I
IClOϩ•NO
␴
due to the bandwidth of the VUV radiation are re-
I
I
3
O( P)
␴
ClOϭ␴O•
,
͑6͒
quired.
IOϩ•NClO
To the best of our knowledge, the present work is the
first experimental determination of the absolute photoioniza-
where both NClO and IClOϩ are negative quantities, since ClO
is consumed by UV photolysis. The photoionization cross
1
tion cross section of atomic chlorine below the S threshold,
I
I
section of atomic oxygen ␴O is well known,44 so that ␴ClO as
which is found at Eϭ16.42 eV.27 Several experimental val-
I
well as ␴ can be precisely determined.
Cl
ues of the photoionization cross section of atomic chlorine
The cation intensities IClOϩ and IOϩ are obtained from
the photolysis mass spectrum shown in Fig. 2͑b͒, which is
recorded at Eϭ14.60 eV. The photoionization cross section
of atomic oxygen O(3P) is smooth at this photon energy.
This is a result of direct photoionization into the Oϩ(4S)
continuum yielding ␴IOϭ3.2Ϯ0.14 Mb.44 The intensity ratio
IClOϩ /IOϩ is found to be 8.8Ϯ0.2. From Eq. ͑6͒ follows
I
␴
have been reported for the 1S continuum. At E
2
Cl( P)
I
ϭ21.218 eV (He-I␣ line͒ ␴
30.8Ϯ5.8 Mb,29 33.3Ϯ4.2 Mb,30 and 38.5Ϯ3.1,31 respec-
has been determined to be
2
Cl( P)
tively. The data in Ref. 31 extend toward longer wave-
I
lengths, where ␴Cl( P)ϭ43.6Ϯ3.5 Mb is found at Eϭ16.42
2
I
eV. Moreover, we note that the absolute values of ␴
that
2
Cl( P)
were reported in earlier experimental work were derived
from the average of theoretical values ͑see Ref. 27͒.
The photoionization cross section of chlorine in the 3p4
I
␴
ClOϭ28Ϯ2 Mb.
I
This value is in agreement with recent results, where ␴
ClO
ϭ27Ϯ5 Mb was determined at Eϭ14.60 eV.17 We note that
the earlier experiments were carried out by photolyzing chlo-
rine dioxide ͑OClO͒, according to OClO(2B1)ϩh␯
continua has been calculated before. Hartree–Fock calcula-
I
tions yield the photoionization cross section ␴Cl( P)ϭ48 Mb
2
at Eϭ15.03 eV.56 More recent work has shown that the mix-
ing of different LS-coupled core states of the same total an-
gular momentum Jc has to be considered.57 The total direct
→ClO(2
͟
)ϩO(3P). It is known that nascent ClO from the
photolysis of OClO is rotationally cold, but vibrationally
excited.14 Therefore, the photoionization cross section of vi-
brationally excited ClO was determined in earlier experi-
ments, where a correction for vibrationally cold ClO was
included.17 In contrast, ClO is formed in the present experi-
ments in a thermal equilibrium at room temperature. The
1
photoionization cross section in the D continuum exceeds
30 Mb, according to this work. A comparison of the present
results with those from theoretical work indicates that the
experimental values in the 1D continuum are by Ϸ20%
smaller than those from previous high level calculations.58
This discrepancy cannot be easily explained. Therefore, we
have performed additional experiments in order to confirm
the present experimental results. Molecular chlorine (Cl2) is
photolyzed ␭ϭ308 nm, where the photoabsorption cross sec-
vibrational constant of ClO(X(2
only 1.6% of the ClO is vibrationally excited in the ϭ1
state at Tϭ300 K. Thus the present results correspond to the
first reliable determination of ␴
͟ , so that
)) is 853.8 cmϪ1 55
v
Љ
I
at room tempera-
ClO(v ϭ0)
Љ
tion (␴ 2ϭ0.15 Mb͒ is continuous.59 Chlorine is well-
Cl
ture.
known to undergo at ␭ϭ308 nm photodissociation according
to Ref. 60:
The photolysis mass spectrum shown in Fig. 2͑b͒ is re-
corded at Eϭ14.60 eV. This energy is right above to the
1
onset of the D continuum of Clϩ, which is found at 14.41
ϩ
g
1
2
eV.26 The intensity ratio IClϩ /IOϩ is 8.42Ϯ0.2 at Eϭ14.60
eV ͓cf. Fig. 2͑b͔͒. This leads, similar to Eq. ͑6͒, to the photo-
Cl X ͚ ͒͒ϩh␯→2Cl P͒.
͑ ͑
͑7͒
͑
2
Photoionization of the primary photoproducts Cl(2P) has
been carried out at Eϭ15.12 eV. There are no close-lying
autoionization resonances, so that the experimental photoion-
ization cross section corresponds to that of the Clϩ(3P
ϩ1D) continuum. Photoionization of molecular chlorine has
been investigated before.61 The appearance energy of Clϩ is
found at Eϭ15.50 eV, so that ionic fragmentation of Cl2 into
ClϩϩCl cannot occur at Eϭ15.12 eV. The photoionization
I
ionization cross section of Cl(2P): ␴Clϭ27Ϯ2 Mb, where
␴IOϭ3.2Ϯ0.14 Mb is used ͑cf. Ref. 44͒. We note that auto-
ionization resonances occur in the photoionization cross sec-
tion of atomic chlorine near Eϭ14.60 eV.27 These are due to
Cl(2P)→3p4(1S0)3d and Cl(2P)→3p4(1S0)5s transitions.
They are expected to increase the Clϩ intensity as a result of
the considerable bandwidth of the VUV radiation. We have
convoluted the Clϩyield from Ref. 27 with the bandwidth of
the VUV radiation in order to account for possible enhance-
ments of the Clϩ signal as a result of autoionization. The
results indicate that autoionization enhances the cation signal
at the present spectral resolution by 12% compared to the
corresponding cross section at 14.60 eV, which is due to
photoionization into the (3Pϩ1D) continuum.27 Therefore,
we correct the photoionization cross section of Cl(2P) at
14.60 eV to
cross section of molecular chlorine at Eϭ15.12 eV has been
I
reported before: ␴ ϭ68.7 Mb.61
Cl
2
The photolysis mass spectrum of chlorine is shown in
Fig. 6. The mass spectrum clearly shows evidence for the
photolysis of molecular chlorine yielding Cl, which is probed
by photoionization mass spectrometry as Clϩ. The intensity
ratio of the Clϩ/Cl2ϩ signals is 0.67Ϯ0.06, where the error
limit is estimated from the statistical error of several repeated
experiments. Considering the stoichiometry of the photolysis
I
␴
ϭ24Ϯ2 Mb.
2
Cl͑ P͒
process ⌬N(Cl)ϭ2⌬N(Cl ) one obtains from Eq. ͑5͒
͓ ͔
2
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Photoionization of photoproducts of ClO
9669
to an Oϩ signal that almost exclusively probes O(1D). The
experimental intensity ratio is I(Oϩ)/I(Clϩ)ϭ0.31Ϯ0.01.
The estimate of the photoionization cross section of O(1D)
requires, as a result of the bandwidth of the VUV radiation,
one to take into account contributions from autoionization of
Cl(2P) as well as O(1D). We estimate that the Clϩ intensity
is enhanced by Ϸ17% due to close-lying autoionization reso-
nances, whereas the Oϩ intensity is enhanced by Ϸ12% for
the same reason. These results are used for correcting the
experimental Oϩ/Clϩ ratio. The photoionization cross sec-
tion of Cl(2P) at 15.4 eV is 24Ϯ2 Mb, according to the
present work. This allows us to derive the absolute photoion-
ization cross section of O(1D) at 15.41 eV:
IO
ϩ
1.17
1.12
I
␴
ϭ
•
• 24Ϯ2 Mb͒ϭ7.8Ϯ1 Mb.
͑
1
O͑ D͒
FIG. 6. Photolysis mass spectrum of Cl2 . Experimental conditions: primary
photolysis at ␭ϭ308 nm, photoionization at Eϭ15.12 eV.
ICl
ϩ
This is to the best of our knowledge the first experimental
I
value of ␴O( D) . The absolute photoionization cross section
1
I
2
of O(1D) has been calculated above the D ionization limit
␴
Cl( P)ϭ23Ϯ2 Mb at Eϭ15.12 eV. This result is in excel-
2
I
(EϾ14.975 eV62͒, where ␴O( D)Ϸ7.0 Mb63 and 7.3Ϯ0.4
lent agreement with the results obtained from the photolysis
of ClO, where ␴Cl( P)ϭ24Ϯ2 Mb is found at Eϭ14.60 eV
1
I
Mb64 are reported at Eϭ15.41 eV. These results are in agree-
2
͑see above͒, considering that the photoionization cross sec-
ment with the present results. The agreement between the
tion remains constant at both photon energies.27
theoretical and experimental photoionization cross section of
I
O(1D) also confirms the reliability of ␴
.
The photolysis of ClO leads to the formation of
2
Cl( P)
Cl(2P3/2) and O(1D) above the threshold of the A(2
͟) con-
tinuum ͓see Eq. ͑2͔͒. It has been shown previously that the
photoionization cross section of Cl(2P) is derived from
IV. CONCLUSIONS
photoionization cross section of O(3P). Similarly, the photo-
I
Cl
ionization cross section of O(1D) can be obtained from ␴
.
The photolysis of ClO is investigated in the
͟)-excitation regime. Single-photon ionization is used to
The determination of the absolute photoionization cross sec-
tion of O(1D) is accomplished by tuning the VUV photon
energy to 15.41 eV ͑see Fig. 7͒. This energy is within the
A(2
detect the primary photofragments. The predissociative for-
mation of Cl(2P)ϩO(3P) is found in the bound region of
the A state. Cl(2P)ϩO(1D) production is detected above the
2
first optically allowed D continuum of O(1D), where con-
tributions from autoionization are small. Therefore, the ex-
perimental results are not strongly affected by the bandwidth
of the VUV radiation. Photolytic formation of O(1D) from
ClO occurs almost quantitatively above the Cl(2P)
ϩO(1D) threshold near 263 nm,7,23 where Ϸ3% of O(3P)
are formed at ␭ϭ248 nm.11 Therefore, photolysis of ClO at
␭ϭ248 nm and subsequent photoionization at 15.41 eV leads
dissociation limit. The threshold of O(1D) formation is
Ϫ1
˜
found at ␭ϭ263.71Ϯ0.01 nm (␯ϭ37 920Ϯ2 cm ͒. The on-
set region of the O(1D) yield is governed by a steep onset,
where low-lying rotational levels of the room temperature
ClO sample lead to O(1D) formation. Higher rotational lev-
els do not contribute to the onset of the Oϩ yield, which is
assigned to a rotational barrier. Thermal excitation of the
sample leads to a small redshift of 106Ϯ9 cmϪ1 between the
experimental data reported in this work and the threshold
value from optical data ͑cf. Ref. 47͒.
The absolute photoionization cross sections of
ClO(X(2
͟
,
ϭ0)), Cl(2P), and O(1D) are determined at
v
Љ
selected VUV photon energies. The reliability of the Cl(2P)
photoionization cross section, that is obtained from the pho-
tolysis of ClO, is confirmed by additional experiments,
where molecular chlorine is photolyzed. We note that recent
theoretical work on Cl(2P) gives a photoionization cross
section that is by Ϸ20% larger than the present experimental
results. In contrast, the photoionization cross section of
O(1D) is in agreement with previous calculations.
The present study shows that reliable and consistent re-
sults on the photolysis of atmospheric trace gases are ob-
tained from photolysis experiments that are followed by
single-photon ionization, similar to our previous works.16,39
This will allow us to apply this approach to other atmo-
FIG. 7. Photolysis mass spectrum of ClO. Experimental conditions: primary
photolysis at ␭ϭ248 nm, photoionization at Eϭ15.41 eV.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06

9670
J. Chem. Phys., Vol. 117, No. 21, 1 December 2002
Flesch et al.
spheric trace gas species, where the primary photoproducts
have been the subject of controversy, such as halogen
nitrates18 and the ClO-dimer.65,66
33 D. R. Flower, in Atoms in Astrophysics, edited by P. G. Burke, W. B.
Eissner, D. G. Hummer, and I. C. Percival ͑Plenum, New York, 1983͒, p.
289.
34 J. B. Burkholder, J. J. Orlando, and C. J. Howard, J. Phys. Chem. 94, 687
͑1990͒.
35 R. I. Derby and W. S. Hutchinson, Inorg. Synth. 4, 152 ͑1953͒.
ACKNOWLEDGMENTS
36 V. H. Diebeler, J. A. Walker, K. E. McCulloh, and H. M. Rosenstock, Int.
One of the authors ͑J.P.͒ acknowledges financial support
by the Deutsche Bundesstiftung Umwelt. Financial support
J. Mass Spectrom. Ion Phys. 7, 209 ͑1971͒.
37
¨
¨
R. Flesch, E. Ruhl, K. Hottmann, and H. Baumgartel, J. Phys. Chem. 97,
837 ͑1993͒.
¨
by the Bundesministerium fur Bildung und Forschung
38 M. Schwell, H.-W. Jochims, B. Wassermann, U. Rockland, R. Flesch, and
͑BMBF͒ and by the Fonds der Chemischen Industrie are
gratefully acknowledged.
¨
E. Ruhl, J. Phys. Chem. 100, 10070 ͑1996͒.
39
¨
R. Flesch, M. C. Schurmann, J. Plenge, H. Meiss, M. Hunnekuhl, and E.
¨
Ruhl, Phys. Rev. A 62, 052723 ͑2000͒.
40
1 R. P. Wayne, Chemistry of Atmospheres ͑Oxford University Press, Oxford,
2000͒.
R. Flesch, J. Plenge, M. C. Schurmann, S. Kuhl, M. Klusmann, and E.
¨
¨
¨
Ruhl, Surf. Rev. Lett. 9, 105 ͑2002͒.
2 J. G. Anderson, W. H. Brune, S. A. Lloyd, D. W. Toohey, S. P. Sander,
W. L. Starr, M. Loewenstein, and J. R. Podolske, J. Geophys. Res.,
͓Atmospheres͔ 94, 11480 ͑1989͒.
41 P. M. Dehmer, J. Berkowitz, and W. A. Chupka, J. Chem. Phys. 59, 5777
͑1973͒.
42 M. Mandelman and R. W. Nicholls, J. Quant. Spectrosc. Radiat. Transf.
17, 483 ͑1977͒.
3 I. A. MacKenzie, R. S. Harwood, L. Froidevaux, W. G. Read, and J. W.
Waters, J. Geophys. Res., ͓Atmospheres͔ 101, 14505 ͑1996͒.
4 S. Solomon, Rev. Geophys. 37, 275 ͑1999͒.
5 J. Wu and A. E. Dessler, J. Geophys. Res., ͓Atmospheres͔ 106, 3195
͑2001͒.
43 S. G. Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D. Levin, and
W. G. Mallard, in NIST Chemistry WebBook, NIST Standard Reference
Database 69, edited by W. G. Mallard and P. J. Linstrom ͑NIST, Gaith-
ersburg, 2001͒.
6 E. Bishenden, J. Haddock, and D. J. Donaldson, J. Phys. Chem. 95, 2113
͑1991͒.
44 G. C. Angel and J. A. R. Samson, Phys. Rev. A 38, 5578 ͑1988͒.
45 F. G. Simon, W. Schneider, G. K. Moortgat, and J. P. Burrows, J. Photo-
chem. Photobiol., A 55, 1 ͑1990͒.
7 S. Schmidt, T. Benter, and R. N. Schindler, Chem. Phys. Lett. 282, 292
͑1998͒.
46 N. Basco and R. D. Morse, J. Mol. Spectrosc. 45, 35 ͑1973͒.
47 P. W. McLoughlin, C. R. Park, and J. R. Wiesenfeld, J. Mol. Spectrosc.
162, 307 ͑1993͒.
8 P. Zou, H. Kim, and S. W. North, J. Chem. Phys. 116, 4176 ͑2002͒.
9
¨
S. Baumgartel and K.-H. Gericke, Chem. Phys. Lett. 227, 461 ͑1994͒.
10 M. Roth, C. Maul, and K.-H. Gericke, J. Chem. Phys. 107, 10582 ͑1997͒.
11 H. F. Davis and Y. T. Lee, J. Phys. Chem. 100, 30 ͑1996͒.
48 R. J. LeRoy and G. T. Kraemer, Chem. Phys. Res. Report CP-650-R,
University of Waterloo, Waterloo, Ont., Canada, 2001.
49 S. A. Barton, J. A. Coxon, and U. K. Roychowdhury, Can. J. Phys. 62, 473
͑1984͒.
12 A. Furlan, M. A. Haeberli, and J. R. Huber, J. Phys. Chem. A 104, 10392
͑2000͒.
13
¨
R. Flesch, B. Wassermann, B. Rothmund, and E. Ruhl, J. Phys. Chem. 98,
50 J. O. Arnold, E. E. Whiting, and S. R. Langhoff, J. Chem. Phys. 66, 4459
͑1977͒.
6263 ͑1994͒.
14
15
¨
E. Ruhl, A. Jefferson, and V. Vaida, J. Phys. Chem. 94, 2990 ͑1990͒.
51 J. A. Coxon, W. E. Jones, and E. G. Skolnik, Can. J. Phys. 54, 1043
͑1976͒.
¨
J. B. Pallix, U. Schuhle, C. H. Becker, and D. L. Huestis, Anal. Chem. 61,
805 ͑1989͒.
52 S. T. Gibson, B. R. Lewis, K. G. H. Baldwin, and J. H. Carver, J. Chem.
Phys. 94, 1060 ͑1991͒.
16
17
¨
¨
J. Plenge, R. Flesch, M. C. Schurmann, and E. Ruhl, J. Phys. Chem. A
105, 4844 ͑2001͒.
R. Flesch, M. C. Schurmann, J. Plenge, M. Hunnekuhl, H. Meiss, M.
53 D. H. Burde, R. A. McFarlane, and J. R. Wiesenfeld, Phys. Rev. A 10,
1917 ͑1974͒.
¨
¨
Bischof, and E. Ruhl, Phys. Chem. Chem. Phys. 1, 5423 ͑1999͒.
54 A. Balakrishnan, V. Smith, and B. P. Stoicheff, Phys. Rev. Lett. 68, 2149
͑1992͒.
18 R. P. Wayne, G. Poulet, P. Biggs et al., Atmos. Environ. 29, 2675 ͑1995͒.
19 G. Porter, Proc. R. Soc. London, Ser. A 200, 284 ͑1950͒.
20 J. A. Coxon and D. A. Ramsay, Can. J. Phys. 54, 1034 ͑1976͒.
21 I. C. Lane, W. H. Howie, and A. J. Orr-Ewing, Phys. Chem. Chem. Phys.
1, 3087 ͑1999͒.
55 K. P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure,
IV. Constants of Diatomic Molecules ͑Van Nostrand, New York, 1979͒.
56 S. T. Manson, A. Msezane, A. F. Starace, and S. Shahabi, Phys. Rev. A 20,
1005 ͑1979͒.
22 A. Toniolo, M. Persico, and D. Pitea, J. Chem. Phys. 112, 2790 ͑2000͒.
23 R. A. Durie and D. A. Ramsay, Can. J. Phys. 36, 35 ͑1958͒.
24 M. Kutzner, in VUV and Soft X-Ray Photoionization, edited by U. Becker
and D. A. Shirley ͑Plenum, New York, 1996͒, p. 47.
57 F. Robicheaux and C. H. Greene, Phys. Rev. A 46, 3821 ͑1992͒.
58 S. S. Tayal, Phys. Rev. A 47, 182 ͑1993͒.
59 D. Maric, J. P. Burrows, R. Meller, and G. K. Moortgat, J. Photochem.
Photobiol., A 70, 205 ͑1993͒.
25 L. J. Radziemski and V. Kaufman, J. Opt. Soc. Am. 59, 424 ͑1969͒.
26 R. L. Kelly, J. Phys. Chem. Ref. Data 16, 1 ͑1987͒.
60 J. G. Calvert and J. Pitts, Photochemistry ͑Wiley, New York, 1966͒.
61 J. A. R. Samson and G. C. Angel, J. Chem. Phys. 86, 1814 ͑1987͒.
62 R. E. Huffman, J. C. Larrabee, and Y. Tanaka, J. Chem. Phys. 47, 4462
͑1967͒.
27 B. Ruscic and J. Berkowitz, Phys. Rev. Lett. 50, 675 ͑1983͒.
28 B. Ruscic and J. Berkowitz, Phys. Rev. A 40, 6716 ͑1989͒.
29 P. van der Meulen, M. O. Krause, C. D. Caldwell, S. B. Whitfield, and C.
A. de Lange, Phys. Rev. A 46, 2468 ͑1992͒.
63 K. L. Bell, P. G. Burke, A. Hibbert, and A. E. Kingston, J. Phys. B 22,
3197 ͑1989͒.
30 W. J. van der Meer, R. Butselaar, and C. A. de Lange, Aust. J. Phys. 39,
779 ͑1986͒.
64 K. Butler and C. J. Zeippen, Astron. Astrophys. 234, 569 ͑1990͒.
65 M. Molina, A. J. Colussi, L. T. Molina, R. N. Schindler, and T.-L. Tso,
Chem. Phys. Lett. 173, 310 ͑1990͒.
31 J. A. R. Samson, Y. Shefer, and G. C. Angel, Phys. Rev. Lett. 56, 2020
͑1986͒.
66 T. A. Moore, M. Okumura, J. W. Seale, and T. K. Minton, J. Phys. Chem.
103, 1691 ͑1999͒.
32
¨
R. Flesch, M. C. Schurmann, M. Hunnekuhl, H. Meiss, J. Plenge, and E.
¨
Ruhl, Rev. Sci. Instrum. 71, 1319 ͑2000͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.114.34.22 On: Sat, 22 Nov 2014 01:58:06