The photodissociation of ClNO2 in argon and water clusters studied at 235 nm by the REMPI-TOF method

Article abstract of DOI:10.1016/S0009-2614(02)00120-3

The photodissociation of ClNO₂ embedded in argon and water clusters was investigated at 235 nm by detecting the Cl(²P_3/2) photofragments with resonance enhanced multiphoton ionization time-of-flight spectroscopy (REMPI-TOF). Under various cluster formation conditions the speed distributions of Cl(²P_3/2) were recorded. In contrast to the monomer photolysis only one decay channel was found to be active showing the fragments to be Boltzmann distributed and isotropic. In view of atmospheric chemistry, CINO₂ in water clusters is a photolytical source of Cl and NO₂ radicals with low translational and internal energy. They are thus less favorable for consecutive reactions than the monomer products.

Full text of DOI:10.1016/S0009-2614(02)00120-3

5 March 2002
Chemical Physics Letters 354 (2002) 120–127
www.elsevier.com/locate/cplett
The photodissociation of ClNO2 in argon and water
clusters studied at 235 nm by the REMPI-TOF method
*
Qiang Li, J. Robert Huber
Physikalisch-Chemisches Institut der Universit€at Z€urich, Winterthurerstrasse 190, CH-8507 Z€urich, Switzerland
Abstract
The photodissociation of ClNO2 embedded in argon and water clusters was investigated at 235 nm by detecting the
2
Clð P3=2Þ photofragments with resonance enhanced multiphoton ionization time-of-flight spectroscopy (REMPI-TOF).
2
Under various cluster formation conditions the speed distributions of Clð P3=2Þ were recorded. In contrast to the
monomer photolysis only one decay channel was found to be active showing the fragments to be Boltzmann distributed
and isotropic. In view of atmospheric chemistry, ClNO2 in water clusters is a photolytical source of Cl and NO2 radicals
with low translational and internal energy. They are thus less favorable for consecutive reactions than the monomer
products. Ó 2002 Elsevier Science B.V. All rights reserved.
1. Introduction
provide Cl and NO2 [7–9]. It has therefore been
suggested that ClNO2 plays a role in photochem-
ical air pollution in the marine troposphere [1,10]
and in catalytic ozone depletion in the stratosphere
[2,3,11,12].
The UV spectrum of ClNO2 at room tempera-
ture starts with a weak, unstructured absorption
between 300 and 400 nm followed by an onset at
around 280 nm, a peak at 215 nm ðr ¼ 3:2 Â
10À18 cm2Þ and a strong increase below 200 nm
[7,8,13]. For the present study we used an excita-
tion wavelength of 235 nm ðr ¼ 1:5 Â 10À18 cm2
[13]) which corresponds to 508 kJ/mol. Under
these conditions the following reaction channels
are energetically accessible according to thermo-
chemical data [14].
Nitryl chloride, ClNO2, is a molecule of atmo-
spheric interest. It is formed by heterogeneous re-
actions of gaseous N2O5 with HCl on ice surfaces
in the stratospheric clouds and by N2O5(g) with
NaCl(s) in the marine boundary layer [1–3]. Both
of these reactions are significant in atmospheric
chemistry because they convert inert solid chloride
to a gas phase active chlorine compound. Re-
cently, the emission inventories for major reactive
chlorine species have been explored and based on a
general circulation model the concentration of
ClNO2 has been calculated [4,5]. Behnke et al. [6],
studying the interactions between N2O5 and NaCl
in an aerosol smog chamber, proposed a forma-
tion mechanism for ClNO2. Exposed to solar ra-
diation this product is readily photolysed to
ClNO2 þ hm ! Cl þ NO2
ð1Þ
DH ¼ 138 kJ=mol
* Corresponding author. Fax: +41-1-6356838.
E-mail address: jrhuber@pci.unizh.ch (J. Robert Huber).
! ClOþNO DH ¼ 174 kJ=mol ð2Þ
0009-2614/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved.
PII: S0009-2614(02)00120-3

Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
121
revealed the NO2 fragments to be produced not
only in the electronic ground state X2A1 ðꢀ 30%Þ
and the excited electronic states A2B2 ðꢀ 30%Þ,
but probably also in the B2B1 state ðꢀ 40%Þ.
Very recently, Plenge et al. [20] reported results
on the photofragmentation of ClNO2 at 240 and
308 nm employing photoionization mass spect-
rometry. For the two important channels (1) and
(3) they found /ðClÞ ¼ 0:93 Æ 0:10 and /ðOÞ ¼
0:07 Æ 0:01, respectively after 308 nm excitation.
However, at 240 nm the Cl production was found
to be significantly reduced relative to that of oxy-
gen ð/ðClÞ=/ðOÞ ¼ 1:44 Æ 0:15Þ which could not
be conciliated with previous results. At both pho-
tolysis wavelengths the atomic oxygen was formed
! ClNOþO DH ¼ 283 kJ=mol ð3Þ
!ClþOþNO DH ¼439kJ=mol ð4Þ
The reaction enthalpies refer to the products in
their electronic ground states. However, it is pos-
sible that the photofragments are formed in elec-
tronically excited states which is relevant for the Cl
and O atoms having low lying spin-orbit excited
states and also for the NO2 molecule where the
A2B2; B2B1 and C2A2 states lie at 9726, 14 743 and
16360 cmÀ1 above the X2A1 ground state, re-
spectively [15–17].
To the best of our knowledge, there is no evi-
dence for the rearrangement of the molecule
leading to ClO + NO (channel 2) and for the three-
body decay process to Cl + O + NO (channel 4)
[13]. The photodissociation of ClNO2 at 350 nm
has been studied earlier by Johnston and co-
workers using resonance fluorescence detection of
Cl and O atoms [8]. Dissociation of the N–Cl bond
to yield Cl þ NO2 (channel 1) was found to be the
dominant process with a quantum yield of
/ðClÞ ¼ 0:93 Æ 0:15; the decay path to ClNO + O
(channel 3) was of minor importance ð/ðOÞ <
0:02Þ. In addition, after excitation at 308 and 235
nm they observed a broad emission from elec-
tronically excited NO2 fragments, but could not
identify the emitting states [18]. Recently Carter
et al. [19] investigated the 235 nm photodissocia-
tion of ClNO2 by the REMPI-TOF detection
method. A bimodal velocity distribution of the
3
in its P2 ground state.
Owing to the potential importance of the
photodissociation of nitric acid, HONO2, attached
to ice particles in the stratosphere [21–23] we
studied water-clustered HONO2 at 193 nm [24].
This photolysis wavelength reproduces well the
solar photolysis conditions in the stratosphere,
where the 190–230 nm region is largely unfiltered
by O3; O2 and N2. In the same spirit we explored
ClNO2 (an isovalent molecule of HONO2) and
report here the results of the 235 nm photodisso-
ciation of ClNO2 attached to or embedded in Ar
and H2O clusters with an average cluster size
n ꢀ 200–800. Using resonance enhanced multi-
photon ionization time-of-flight spectroscopy
(REMPI-TOF) the velocity distribution of the
2
Clð P3=2
Þ
fragments emerging from clustered
2
nascent Cl atoms in the P3=2 ground state was
observed which led to the conclusion that two
decay channels are active in reaction (1):
ClNO2 was investigated and compared to results
previously obtained from ClNO2 monomer
photolysis.
ClNO2 þ hm ! Cl þ NO2ðX2A1Þ
! Cl þ NO2ðA2B2Þ
ð1aÞ
ð1bÞ
2. Experiment
The minor reaction leads to NO2ðX2A1Þ with
/ ¼ 0:15 Æ 0:05, the dominant one to NO2ðA2B2Þ
with / ¼ 0:85 Æ 0:05. Using photofragment tran-
slation energy spectroscopy (PTS) [13] at 248 nm
this study was extended confirming the formation
of NO2 fragments in the first electronically excited
state A2B2. Furthermore the internal energy dis-
tribution of the NO2 fragments obtained from the
well-structured translational energy distribution
The apparatus has been described in some detail
elsewhere [25,26]. Briefly, it consists of a molecular
beam source chamber, pumped by two diffusion
pumps, and a main chamber, separated by a
skimmer and equipped with a time of flight mass
spectrometer (TOF-MS). The main chamber is
pumped by two turbomolecular pumps, one
mounted directly below the MS accelerator stage
and one at the detector end of the flight tube. The

122
Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
TOF-MS is a two stage Wiley–McLaren design
with an extraction and an acceleration region [27].
Two orthogonal pairs of deflection plates allow
any drift in the ion trajectories to be corrected. The
photofragment ions are detected by a multichannel
plate detector (Galilleo FT 4000) which is moun-
ted at the end of a 40 cm long field-free flight tube.
The ion signal is amplified by a 500 MHz pream-
plifier and digitized by a 500 MHz digital sampling
oscilloscope (HP5422A). A sufficient signal to
noise ratio for the TOF measurements is typically
achieved after about 500 shots.
In the case of water cluster formation, a mixture of
0.4% ClNO2 in 2 bar He was guided into a flask
filled with H2O at temperature between 30 and
65 °C. The outgoing gas mixture, possessing a
partial pressure of pðH2OÞ ¼ 40–250 mbar, was
then passed through a heated tube to the pulsed
valve. The latter was also heated to prevent con-
densation of water. Based on the work of Buck
and Krohne [30] and Hagena [31], the average
cluster size for the argon and the water clusters is
expected to be in the range of ꢀ200–400 and
ꢀ400–800, respectively.
The photodissociation experiments were car-
ried out with a Lambda Physik FL2002 dye laser
pumped by a Lambda Physik LPX200 excimer
laser running on XeCl at 308 nm. The laser beam
is focused by a 40 cm focal length lens into the
middle of the extraction region of the TOF-MS
where it crosses the molecular beam. In the pre-
sent work, the dye laser pulse ran on Coumarin
47 and the output was frequency doubled with a
BBO crystal yielding a pulse energy of 400 lJ
3. Results
2
The TOF profiles recorded for the Clð P3=2
Þ
fragments following photodissociation of ArnClNO2
and ðH2OÞnClNO2 clusters at 235.27 nm are dis-
played in Fig. 1a and b for the laser polarization
parallel, at the magic angle and perpendicular to
the detection axis, respectively. The feature on the
right-hand side of the profiles is due to the onset of
the corresponding TOF profile of the 37Cl isotope.
The analysis by a forward convolution proce-
dure [26] yielded three speed distributions PðvÞ of
Cl photofragments indicated as dotted, dashed
and chain-dashed lines. These distributions are
shown in Fig. 2a and b for the ArnClNO2 and
ðH2OÞnClNO2 clusters, respectively. The main,
middle component (dashed line) consists of ꢀ62%
and 60% of all the detected Cl fragments from the
ArnClNO2 and the (H2OÞn ClNO2 clusters, re-
spectively, and is centered at ꢀ1600 m/s for both
distributions. The faster component (chain-dashed
lines) consists of 14% and 12% of the detected Cl
fragments from ArnClNO2 and ðH2OÞnClNO2,
respectively, and has a mean speed of ꢀ2700 m/s
for both clusters. The analysis further revealed
that the two components have the same anisotropy
parameter b ¼ þ1:0 Æ 0:1 in both types of cluster
dissociations. The ratio between the two compo-
nents is thus about 82:18. Based on the earlier
findings of the monomer photodissociation per-
formed at the same excitation wavelength [19], we
assign the fast component of the Cl fragment dis-
tributions to reaction (1a) with the products
Cl þ NO2 ðX2A1Þ and the middle component to
around 235 nm with a bandwith of 0.4 cmÀ1
.
Measurements on the Cl photofragments were
performed using a one-colour arrangement in
which the same dissociation laser pulse was used
for both dissociation and detection. The Cl2 P3=2
atomic ground state was probed by (2 þ 1) RE-
o
o
2
2
MPIat 235.27 nm via the 4p D3=2 3p5 P3=2
transition [28]. The dissociation laser was linearly
polarized to ꢀ97% and the polarization could be
rotated by a suitable k/2 plate. The photodisso-
ciation experiments were carried out with parallel
(v ¼ 0°), perpendicular (v ¼ 90°), and magic
angle (v ¼ 54:7°) polarizations relative to the de-
tection axis. The photofragment speed distribu-
tions PðvÞ were obtained from the TOF data using
a forward convolution procedure as previously
described [26].
Nitryl chloride was synthesized by passing HCl
gas through a mixture of nitric and sulfuric acid
and collecting the product in a low-temperature
bath [29]. In order to generate the cluster beam, a
conical nozzle (aperture d ¼ 1 mm, length 10 mm,
opening angle 2h ¼ 20°) mounted on a piezoelec-
tric pulsed valve was used. Ar cluster with ClNO2
was generated by expanding a premixture of 0.5–
2% ClNO2 in Ar at a backing pressure of 2 bar Ar.

Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
123
2
Fig. 1. REMPI-TOF spectra of Clð P3=2Þ fragments from ClNO2 photodissociation at 235 nm recorded under three different polar-
ization angles for (a) argon clusters and (b) water clusters. The expansion conditions were: (a) 2 bar Ar with 0.5% ClNO2, (b) 2 bar He
with 0.4% ClNO2 and 250 mbar H2O. The open circles are the experimental data. The dotted line indicates Cl fragments from cluster
dissociation, the chain-dashed line those from channel 1(a), and the dashed line those from channel 1(b); the soild line is the sum of the
three contributions. The corresponding photofragment speed distributions are given in Fig. 2.
reaction (1b) with the products Cl þ NO2ðA2B2Þ.
For the monomer we previously found the two Cl
speed distributions centered at 2700 and 1700 m/s
and the ratio between the slow and fast component
85:15 [19]. Thus, all these data are the same for
monomer and cluster photolysis within experi-
mental error. On the other hand, the slow com-
ponent (indicated by dotted lines in Fig. 2), which
accounts for 24% and 28% of all detected Cl
fragments, has a mean speed of 590 and 620 m/s
for ArnClNO2 and ðH2OÞnClNO2 clusters, respec-
tively. Both cluster types show the fragment an-
isotropy parameter to be b ¼ 0:0 Æ 0:1: This
component of the Cl products is attributed to
emerge from the photodissociation of ClNO2 in-
corporated in clusters (see below).
In order to explore cluster formation and clus-
ter size distribution, the ClNO2 concentration in
the premixed samples with Ar and H2O was var-
ied. Fig. 3 shows the effect on the TOF profiles of
the Cl photofragments on reducing the ClNO2
concentration in Ar from 2% to 0.5% while keep-
ing the Ar stagnation pressure at 1.5 bar. A dis-
tinct, relative increase of a factor of almost two
(from 11% to 20%) is observed for the slow com-
ponent in Fig. 3a paralleled by a corresponding
increase of the isotropic part shown in Fig. 3b
(center portion). Moreover the distribution of the

124
Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
2
distributions for Clð P3=2Þ photofragment stem-
ming from photodissociation in an expansion of
0.4% ClNO2 in 2 bar He with varying H2O partial
pressures. The fraction of the slow component is
observed to increase with the water partial pres-
sure from 9% at pðH2OÞ ¼ 40 mbar to 28% at 250
mbar. This is accompanied by a decrease in the
mean speed of the slow component which falls
from 700 to 620 m/s and by a decrease of the
distribution width. The mean speed of the middle
and fast component depends only slightly on
(pH2O) as evident from the reduction from 1750
and 2800 m/s at pðH2OÞ ¼ 40 mbar H2O to 1650
and 2700 m/s at pðH2OÞ ¼ 250 mbar H2O, re-
spectively.
The cluster formation of ðH2OÞnClNO2 and
2
ArnClNO2 was recorded by the Clð P3=2Þ signal as
a
function of the stagnation pressure. The
ðH2OÞnClNO2 cluster measurements were carried
out using a mixture of 0.4% ClNO2 in He at a
pressure of 0.4 to 2.5 bar and a constant H2O
partial pressure of 200 mbar. The signal intensity
for the slow component, which exclusively reflects
cluster formation, is depicted in Fig. 5. It shows an
abrupt increase at ꢀ0.9 bar reaching a constant
maximum value at ꢀ1.4 bar. This behavior is
reminiscent of a ‘‘phase transition’’ associated with
cluster formation [25,26]. At a stagnation pressure
below the onset pressure, the TOF profile is bi-
modal paralleling that of the monomer dissocia-
tion while pressures above the threshold yield
trimodal profiles which contain the additional slow
component. A similar behavior was observed for
ArnClNO2 as obtained with a mixture of 0.1%
ClNO2 in Ar and a changing pressure from 0.4 to
2.5 bar. The phase transition occurs here between
ꢀ0.8 and 1.2 bar (cf. Fig. 5).
2
Fig. 2. Clð P3=2Þ photofragment speed distributions PðvÞ from
the REMPI-TOF data in Fig. 1, (a) ArnClNO2 and (b)
ðH2OÞnClNO2. The dotted, dashed and chain-dashed lines in-
dicate Cl fragments form cluster dissociation, from channel 1(b)
and from channel 1(a), respectively. The arrows mark the
maximum thermodynamically allowed speed ETmax for the re-
spective component.
slow Cl fragments is narrowed and its mean ve-
locity shifted from 750 to 590 m/s due to this de-
crease in the ClNO2 concentration. A similar effect
is observed when the stagnation pressure of Ar is
increased. Going from 1.5 to 2.5 bar and keeping
the ClNO2 concentration constant at 0.5%, the
relative increase of the slow component is from
20% to 31% accompanied by a narrowing of its
distribution. In both cases, i.e. for the ClNO2
concentration decrease and the Ar pressure in-
crease, the velocity distributions of the fast and
medium components of the Cl fragments remain
unchanged.
4. Discussion
The photodissociation of ClNO2 monomer at
235 nm has been studied by the REMPI-TOF
2
method detecting the Cl ground state ð P3=2
Þ
photofragment [19]. Two contributions were ob-
2
served in the speed distribution of Clð P3=2Þ,
The corresponding results for the ðH2OÞnClNO2
15 Æ 5% of the fragments centered at 2700 m/s and
85 Æ 5% centered at 1700 m/s. The fast fragments
clusters are displayed in Fig. 4. It shows the speed

Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
125
2
Fig. 3. REMPI-TOF spectra of Clð P3=2Þ photofragments (b) and the corresponding speed distributions PðvÞ (a), as a function of the
ClNO2 concentration seeded in Ar at a stagnation pressure of 1.5 bar. The arrows mark ETmax
.
were attributed to decay channel 1(a) and the slow
ones to 1(b) where the NO2 counterfragment
emerges in the ground X2A1 and electronically
excited A2B2 state, respectively. The recoil anisot-
ropy b was measured to be þ1:1 Æ 0:1, suggesting a
transition dipole moment along the N–Cl bond to
an excited electronic state of A1 symmetry. The
dissociation lifetime estimated from this b value is
shorter than 2.2 ps. At the excitation wavelength of
235 nm corresponding to 508 kJ/mol, the available
energy Eavl obtained by the energy balance
The Cl photofragement speed distribution re-
corded under various cluster formation condi-
tions are shown in Fig. 3 for ArnClNO2, and in
Fig. 4 for ðH2OÞnClNO2 illustrating the influence
of a cluster environment on the photofragmen-
tation of ClNO2. While for the monomer disso-
ciation a bimodal speed distribution of the Cl
fragment has been found [19], we observed three
speed distributions. The analysis revealed that the
fast and the medium component, which show the
same intensity ratio and the same speed maxima
under the various conditions applied, parallel the
behavior of the two components found in the
monomer dissociation for reactions (1a) and (1b).
Furthermore the recoil anisotropy for both of
these components, their non-Boltzmann speed
distributions and the absence of a phase transi-
tion with increasing stagnation pressure are in
excellent agreement with the monomer photolysis
[19]. Thus only the slow speed component in
Figs. 3 and 4 appears to be due to the ClNO2
fragmentation in the cluster environment, while
the majority of the detected Cl fragments stems
Eavl ¼ hm À DoðCl À NO2Þ þ EintðClNO2Þ;
ð1cÞ
is 370 kJ=mol; the dissociation energy Do
ðCl–NO2Þ is 138 kJ/mol [14] and the internal en-
ergy of the parent molecule EintðClNO2Þ in the
supersonic jet can be neglected. As the average
total translational energy hETi was measured to be
ꢀ220 kJ/mol for channel (1a) and ꢀ100 kJ/mol for
channel (1b) [19], the internal energy Evib þ Erot is
thus 150 kJ/mol for NO2ðX2A1Þ and 154 kJ/mol
for NO2ðA2B2Þ. The electronic excitation of the
latter amounts to 116 kJ/mol [15].

126
Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
Fig. 5. Argon and water cluster formation as a function of
the stagnation pressure p0 measured for the slow component in
Fig. 2. The expansion conditions were (a) 0.4% ClNO2 with
200 mbar H2O in helium and (b) 0.1% ClNO2 in argon.
2
Fig. 4. Clð P3=2Þ photofragment speed distributions PðvÞ as a
function of the partial water pressure pðH2OÞ ¼ 40–250 mbar.
The expansion condition were 2 bar He with 0.4% ClNO2. The
arrows mark ETmax
.
fragments. However in our cluster photodissocia-
tion we detect only the Cl fragments which escape
the cluster cage and for the broad cluster size
distribution created under our conditions we ex-
pect only small clusters to be melted upon pho-
tolysis. In large clusters the Cl fragments will
remain caged.
from free or at most very weakly cluster-bound
ClNO2.
The analysis of the slow Cl fragments from
Fig. 2 shows an isotropic recoil and a Boltzmann
velocity distribution with
a
temperature of
Fig. 3 shows the relative portion of Cl cluster
fragments to increase with decreasing ClNO2
concentration (from 2% to 0.5%) i.e. a higher
percentage of ClNO2 molecules becomes incorpo-
rated in clusters. A similar effect is observed when
the Ar pressure is increased. In the latter case this
is ascribed to the decreasing expansion tempera-
ture with increasing stagnation pressure leading to
more favorable cluster formation conditions. For
water clusters (Fig. 4) the increase of the cluster
material by increasing the H2O partial pressure
from 40 to 250 mbar produces more clusters and
hence more Cl cluster fragments relative to those
produced from the monomers.
450 Æ 100 K and 580 Æ 100 K (corresponding to
hETi of ꢀ5.6 kJ/mol and ꢀ7.2 kJ/mol) for the ar-
gon and water clusters, respectively. These findings
allow us to assign the slow component to frag-
ments which are initially generated by photolysis
in the cluster interior. During the escape from the
evaporating cluster cage they are thermalized and
lose their vector correlation. The small hETi values
calculated above (assuming mCl < mCluster) indicate
that most of Eavl ¼ 370 kJ/mol is transferred to the
cluster cage. From inspection of Figs. 2–4 it ap-
pears as if the portion of the Cl cluster fragments
would generally be small compared to all the Cl

Q. Li, J. Robert Huber / Chemical Physics Letters 354 (2002) 120–127
127
[12] D.V. Michelangeli, M. Allen, Y.L. Yung, Geophys. Res.
Lett. 18 (1991) 673.
Acknowledgements
[13] A. Furlan, M.A. Haeberli, J.R. Huber, J. Phys. Chem. A
104 (2000) 10392.
Support by the Schweizerischer Nationalfonds
and by the Alfred Werner Legat is gratefully ac-
knowledged. We thank Rolf Pfister for the syn-
thesis of ClNO2 and PD Dr. P. Willmott for
critically reading the manuscript.
[14] JANAF Themochemical Tables, J. Phys. Chem. Ref. Data
14 (1985) suppl. 1.
[15] B. Kirmse, A. Delon, R. Jost, J. Chem. Phys. 108 (1998)
6638.
[16] J.L. Hardwick, J.C.D. Brand, Chem. Phys. Lett. 21 (1973)
458.
[17] A. Weaver, R.B. Metz, S.E. Bradforth, D.M. Neumark,
J. Chem. Phys. 90 (1989) 2070.
References
[18] C.E. Miller, H.S. Johnston, J. Phys. Chem. 97 (1993) 9924.
[19] R.T. Carter, A. Hallou, J.R. Huber, Chem. Phys. Lett. 310
(1999) 166.
[1] B.J. Finlayson-Pitts, M.J. Ezell, J.N. Pitts Jr., Nature 337
(1989) 241.
€ €
[20] J. Plenge, R. Flesch, M.C. Schurmann, E. Ruhl, J. Phys.
[2] M.A. Tolbert, M.J. Rossi, D.M. Golden, Science 240
(1988) 1018.
Chem. A 105 (2001) 4844.
[21] D. Salcedo, L.T. Molina, M.J. Molina, J. Phys. Chem. A
105 (2001) 1433.
[3] M. Leu, J. Geophys. Res. Lett. 15 (1988) 851.
[4] D.J. Erickson III, C. Seuzaret, W.C. Keene, S.L. Gong,
J. Geophys. Res. 104 (1999) 8347.
[22] M.A. Zondlo, S.B. Barone, M.A. Tolbert, J. Phys. Chem.
A 102 (1998) 5735.
[23] A.E. Waibel, Th. Peter, K.S. Carslaw, H. Oelhaf,
[5] W.
C Keene, M.A.K. Khalil, D.J. Erickson III, A.
McCulloch, T.E. Graedel, J.M. Lobert, M.L. Aucott,
S.L. Gong, D.B. Harper, G. Kleiman, P. Midgley, R.M.
Moore, C. Seuzaret, W.T. Sturges, C.M. Benkovitz, V.
Koropalov, L.A. Barrie, Y.F. Li, J. Geophys. Res. 104
(1999) 8429.
€
G. Wetzel, P.J. Crutzen, U. Poschl, A. Tsias, E. Reimer,
H. Fischer, Science. 283 (1999) 2064.
[24] Q. Li, J.R. Huber, Chem. Phys. Lett. 345 (2001) 415.
[25] C.J. Kreher, R.T. Carter, J.R. Huber, J. Chem. Phys. 110
(1999) 3309.
[6] W. Behnke, C. George, V. Scheer, C. Zetzsch, J. Geophys.
Res. 102 (1997) 3795.
[26] K. Bergmann, R.T. Carter, G.E. Hall, J.R. Huber,
J. Chem. Phys. 109 (1998) 474.
[7] A.J. Illies, G.A. Takacs, J. Photochem. 6 (1976/77) 35.
[8] H.H. Nelson, H.S. Johnston, J. Phys. Chem. 85 (1981)
3891.
[27] W.C. Wiley, I.H. Mclaren, Rev. Sci. Instrum. 26 (1955)
1150.
[28] C.E. Moore, Atomic Energy Levels, vol. 1, in: Circular
467, National Bureau of Standards, Washington, DC,
1949.
[9] A.J. Grimley, P.L. Houston, J. Chem. Phys. 72 (1980)
1471.
[10] F.E. Livingston, B.J. Finlayson-Pitts, Geophys. Res. Lett.
18 (1991) 17.
[29] M. Volpe, H.S. Johnston, J. Am. Chem. Soc. 78 (1956) 3903.
[30] U. Buck, R. Krohne, Phys. Rev. Lett. 73 (1994) 947.
[31] O.F. Hagena, Rev. Sci. Instrum. 63 (1992) 2374.
[11] B.J. Finlayson-Pitts, F.E. Livingston, H.N. Berko, Nature
343 (1990) 622.