Photofragmentation of nitryl chloride in the ultraviolet regime and vacuum ultraviolet regime

Article abstract of DOI:10.1021/jp0044330

Photofragmentation of nitryl chloride (ClNO₂) is reported in the ultraviolet (UV) (λ = 240 nm and λ = 308 nm) and in the vacuum ultraviolet (VUV) regime (55 nm ≤ λ ≤ 110 nm, corresponding to the photon energy range 11.3 eV ≤ E ≤ 22.5 eV), where pulsed radiation is used to excite the neutral molecule in the gas phase. The neutral photolysis products that are formed upon UV photolysis are subsequently probed by photoionization mass spectrometry by using time-correlated tunable laser-produced plasma VUV radiation. UV-pump/VUV-probe experiments allow us to identify two primary photolysis channels at λ = 308 nm: (i) Cl + NO₂ and (ii) O + ClNO. Primary quantum yields for atomic product formation are deduced from photoionization experiments for both channels: γ_{308 nm}(Cl) = 0.93 ± 0.10, and γ_{308 nm}(O) = 0.07 + 0.01. The yield of Cl formation (N(Cl)) is significantly reduced relative to that of O formation (N(O)) at λ = 240 nm, corresponding to a N(Cl)/N(O) ratio of 1.44 ± 0.15. The atomic oxygen is found to be formed in its ³P ground state at both photolysis wavelengths. The present results are compared to earlier work, and atmospheric implications of the present results are briefly discussed. The tunable VUV light source also allows us to perform photoionization mass spectrometry experiments on nitryl chloride without primary photolysis. These experiments yield the first ionization energy of ClNO₂ as well as fragmentation thresholds of ClNO₂⁺.

Full text of DOI:10.1021/jp0044330

4844
J. Phys. Chem. A 2001, 105, 4844-4850
Photofragmentation of Nitryl Chloride in the Ultraviolet Regime and Vacuum Ultraviolet
Regime
J. Plenge, R. Flesch, M. C. Sch u1 rmann, and E. R u1 hl*
Fachbereich Physik, UniVersit a¨ t Osnabr u¨ ck, Barbarastrasse 7, 49069 Osnabr u¨ ck, Germany
ReceiVed: December 7, 2000; In Final Form: March 6, 2001
Photofragmentation of nitryl chloride (ClNO ) is reported in the ultraviolet (UV) (λ ) 240 nm and λ ) 308
2
nm) and in the vacuum ultraviolet (VUV) regime (55 nm e λ e 110 nm, corresponding to the photon energy
range 11.3 eV e E e 22.5 eV), where pulsed radiation is used to excite the neutral molecule in the gas
phase. The neutral photolysis products that are formed upon UV photolysis are subsequently probed by
photoionization mass spectrometry by using time-correlated tunable laser-produced plasma VUV radiation.
UV-pump/VUV-probe experiments allow us to identify two primary photolysis channels at λ ) 308 nm: (i)
Cl + NO and (ii) O + ClNO. Primary quantum yields for atomic product formation are deduced from
2
photoionization experiments for both channels: γ308 nm(Cl) ) 0.93 ( 0.10, and γ308 nm(O) ) 0.07 ( 0.01. The
yield of Cl formation (N(Cl)) is significantly reduced relative to that of O formation (N(O)) at λ ) 240 nm,
3
corresponding to a N(Cl)/N(O) ratio of 1.44 ( 0.15. The atomic oxygen is found to be formed in its P
ground state at both photolysis wavelengths. The present results are compared to earlier work, and atmospheric
implications of the present results are briefly discussed. The tunable VUV light source also allows us to
perform photoionization mass spectrometry experiments on nitryl chloride without primary photolysis. These
+
experiments yield the first ionization energy of ClNO
2
as well as fragmentation thresholds of ClNO
2
.
I. Introduction
ClNO + hν (λ e 852 nm) f Cl + NO2
(1a)
2
Nitryl chloride (ClNO2) occurs as a trace gas in the
troposphere and in the stratosphere, with a calculated tropo-
spheric nighttime mixing ratio of 24.7 ppt at 60° latitude. Its
3
ClNO + hν (λ e 415 nm) f O ( P) + ClNO (1b)
2
1
ClNO + hν (λ e 267 nm) f Cl + NO + O (1c)
2
tropospheric formation mechanism is related to heterogeneous
processes occurring on sea salt aerosol particles at nighttime
1
2
ClNO + hν (λ e 250 nm) f O ( D) + ClNO (1d)
that require the presence of N2O5. ClNO2 is formed in the
2
stratosphere by the reaction of gaseous N2O5 with hydrogen
chloride-ice surfaces on polar stratospheric clouds. The trace
The products of reaction 1c can also be formed by fragmenta-
tion of NO2 (via 1a) or ClNO (via 1b). Other possible channels
that involve a rearrangement of the molecule leading, for
example, to ClO + NO formation are not considered, because
there is, to date, no evidence for such processes to the best of
our knowledge. The photolysis of ClNO2 has been investigated
earlier by resonance fluorescence detection of the atomic
3
gas is known to be efficiently photolyzed by sunlight during
the day, yielding primarily atomic chlorine. The importance of
nitryl halogenides, such as ClNO2 and BrNO2, with respect to
atmospheric photolysis and the tropospheric ozone budget as
well as stratospheric ozone depletion has been discussed
previously.3
,4
6
products. The process yielding Cl + NO2 is dominant at λ )
The photoabsorption cross section of ClNO2 has been
3
50 nm. Its quantum yield γ350 nm(Cl + NO2) ) 0.93 ( 0.15
investigated earlier in the ultraviolet (UV) regime between λ )
(cf. eq 1a) is considerably greater than that of the competing
-7
1
85 nm and λ ) 400 nm.5 Three broad unstructured bands
process γ350 nm(O + ClNO) < 0.02 (cf. eq 1b). The internal
are observed in this wavelength regime. An intense maximum
in absorption cross section is found at λ ≈ 220 nm, and there
is another considerably weaker maximum at λ ≈ 310 nm. Earlier
studies are in general agreement with respect to the overall shape
of the absorption cross section.5 However, the absolute value
of the UV absorption cross section shows significant differences
in the 350 nm regime, which are of the order of a factor of
energy distribution of the molecular photolysis product NO2 has
been investigated at various wavelengths in the UV regime.8
Recent work on the photolysis of ClNO2 has been carried out
at a UV wavelength of λ ) 235 nm, where resonance-enhanced
multiphoton ionization in combination with time-of-flight mass
,6
9
spectrometry was used. The velocity distribution of nascent
2
atomic chlorine in its P3/2 ground state that is produced via eq
∼
1.7.
1
a led to the conclusion that the companion photoproduct NO2
The photolysis channels of nitryl chloride that can occur at
2
is formed in part in its first electronically excited state A˜ ( B2).
This was confirmed in another study of the same group, where
photofragment translational energy spectroscopy was used to
photolysis wavelengths λ > 240 nm are summarized in eq 1
along with the calculated threshold wavelengths, where most
5
products are formed in their electronic ground state:
detect the primary photolysis products of nitryl chloride at λ )
7
2
48 nm. It was found that channel 1a is dominant at this
*
Corresponding author. FAX: +49-541-969 2264. E-mail: ruehl@uos.de
photolysis wavelength, and there is no evidence for other
1
0.1021/jp0044330 CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/01/2001

Photofragmentation of Nitryl Chloride
J. Phys. Chem. A, Vol. 105, No. 20, 2001 4845
2
channels. However, earlier fluorescence work gave different
results at the same photolysis wavelength, where a yield of about
2
pulse was kept as small as possible (<30 mJ/cm ) in order to
avoid further photolysis of the primary photoproducts. It is
considerably lower than in recent work, where the NO product
channel was shown to be massively affected by high laser
0% was ascribed to channel 1c.10 Finally, translational
spectroscopy experiments indicated that there is a small yield
for channel 1b at λ ) 248 nm.1
0,11
7
fluence.
Variations of the quantum yield of Cl formation from the
photolysis of ClNO2 in the UV regime are expected to be of
crucial importance with respect to subsequent processes that
are initiated by atomic chlorine in the atmosphere, such as ozone
losses in the stratosphere.3 Moreover, various competing
photolysis channels can occur at λ > 240 nm (cf. eqs 1a-d).
In the case of atomic oxygen formation, it is also possible that
ClNO2 was introduced into the ionization region of the TOF
via an effusive jet. Photoionization mass spectra were obtained
from single-photon ionization by using dispersed LPP radiation.
Alternatively, undispersed LPP radiation can be used, if the
VUV monochromator is set to zeroth order. The arrival time of
the cations at the microchannel plate cation detector with respect
to the ionizing laser pulse was recorded by using a digital
oscilloscope (LeCroy 9350 C). Photoion yields of mass-selected
cations were obtained by selecting mass channels in the TOF
mass spectrum as a function of the photon energy. The photon
flux at the exit slit of the VUV monochromator was used for
normalizing the photoion yields. It was recorded simultaneously
by a microchannel plate detector. The reliability of normalization
of the cation intensities to the VUV-photon flux has been
confirmed by comparing the normalized partial cation yield
spectra of well-known species to those that have been obtained
,12
1
excited O ( D) is formed (cf. eq 1d) in addition to ground-state
3
oxygen (O ( P), cf. eqs 1b and 1c). This gives the primary
motivation for the present work.
We have earlier shown that the primary photofragments of
molecules and radicals can be probed efficiently and quantita-
tively by single-photon ionization and subsequent mass spec-
trometric detection of the corresponding singly charged
cations.1
3-16
Single-photon ionization has significant advantages
17
compared to multiphoton ionization, because single-photon
ionization cross sections of atoms and small molecules are often
known precisely so that the results can easily be quantified.
This is often impossible in the case of multiphoton ionization
where resonantly excited intermediate states with unknown
transition moments are involved in the ionization process.
Resonant excitations above the first ionization energy lead in
the vacuum ultraviolet (VUV) regime to autoionization. These
processes yield distinct resonances that allow us to probe
selectively the quantum states of the primary atomic and
molecular photoproducts. In particular, laser-produced plasma
14
from synchrotron radiation experiments. The wavelength
calibration of the VUV monochromator has been established
by using autoionization resonances of rare gases, such as argon
19
and neon.
The primary neutral photofragments of electronically excited
nitryl chloride have been detected by using a pump-probe
1
4-16
detection scheme.
The frequency-doubled output of a dye
laser served for primary photoexcitation of ClNO2 in the UV
regime. Alternatively, a XeCl excimer laser (λ ) 308 nm) was
used for photolysis. Photoionization of the primary photolysis
products has been accomplished by time-correlated LPP radia-
tion after a typical delay of ∆t ≈ 100 ns relative to the primary
photolysis pulse.
(LPP) radiation has been shown to be particularly useful for
single photon ionization that often requires tunable VUV
radiation in a wide energy regime.1
4,16
The use of single-photon
ionization approaches also requires detailed knowledge of
photoionization of the molecule (ClNO2) and photofragmenta-
A typical set of mass spectra at a given UV-excitation
wavelength and VUV photon energy consists of (i) photoion-
ization of the ground-state molecule by VUV radiation; (ii) UV
photoexcitation without subsequent VUV photoionization, so
that possible contributions from multiphoton ionization are
detected; and (iii) primary UV excitation followed by subsequent
VUV photoionization. The data analysis is carried out by
subtracting the contributions that occur in the spectra (i) and
(ii) from the pump-probe TOF mass spectrum (iii). The
resulting difference spectrum is termed photolysis mass spectrum
in the following.
+
tion properties of the corresponding molecular cation (ClNO2 )
in the VUV regime. No work on photoionization mass spec-
trometry of ClNO2 has been published to the best of our
knowledge. However, the He(I) photoelectron spectrum of
18
ClNO2 has been reported earlier. Therefore, we also report
on photoionization mass spectrometry of ClNO2 in the VUV
regime by using monochromatic LPP radiation. This light source
has been shown to be a suitable alternative to synchrotron
14,16
radiation as well as other sources of tunable VUV radiation.
II. Experimental Section
The pressure in the ionization region of the TOF was typically
-
5
≈
1 × 10 mbar. The intensity of cation signals that are due
A detailed description of the experimental setup has been
to ionization of primary photofragments has been verified to
depend linearly on the pressure in the ionization region so that
all experiments were carried out under collision-free conditions.
Moreover, the signal intensity was linearly related to the pump
laser intensity, indicating that exclusively one-photon-induced
photolysis processes were investigated.
1
4
published recently. Briefly, it consists of the following major
components: (i) a tunable, excimer-pumped pulsed dye-laser
(Lambda Physik: Compex 201 and Scanmate) equipped with
a frequency doubling unit (BBO(I)-crystal) for UV excitation
of the gas-phase molecules under collision-free conditions; (ii)
a pulsed, tunable VUV light source that delivers intense,
monochromatic LPP radiation in the VUV regime 8 eV e E e
Nitryl chloride was prepared by the reaction of hydrogen
chloride with dinitrogen pentoxide at T ) 213 K according to
HCl + N2O5 f ClNO2 + HNO3. The HCl was of commercial
quality (Aldrich, purity >99%), N2O5 was prepared by the
2
5 eV, corresponding to 50 e λ e 155 nm. A solid tungsten
target has been used for the efficient production of tunable LPP
radiation. The radiation consists essentially of a broad back-
ground continuum that is dispersed by a 1 m vacuum mono-
chromator (McPherson, Nova 225) equipped with a 1200 l/mm
holographically ruled grating (Jobin Yvon); and (iii) a time-of-
flight mass spectrometer (TOF) for cation separation and
2
0
reaction of ozone with nitrogen dioxide. The product ClNO2
was purified from N2O5 and nitric acid by distillation at T )
213 K. Another product of the synthesis that occurs with a minor
yield of ≈2% was Cl2, which forms an azeotropic mixture with
1
4,16
21
detection.
The bandwidth of the LPP radiation is estimated
ClNO2. The molecular chlorine impurity has not been removed
to be of the order 0.4 nm. The fluence for primary photolysis
from the sample. The purity of the ClNO2 was established by

4846 J. Phys. Chem. A, Vol. 105, No. 20, 2001
Plenge et al.
Figure 1. Time-of-flight mass spectrum of nitryl chloride (ClNO
2
).
Experimental conditions: photoionization with undispersed LPP radia-
Figure 2. Photoion yields of cations that are formed by photoionization
-
5
+
2
tion; p ) 1 × 10 mbar, T ) 300 K. The weak signal of Cl
is due
of ClNO . The spectra are scaled by their relative abundance in
2
to chlorine impurities.
photoionization mass spectrometry. Arrows indicate the threshold
energies.
infrared absorption spectroscopy,22,23 indicating that there were
no significant impurities of HNO3 and ClNO in the sample.
+
occurs at the ionization threshold: ClNO2 + hν f NO2 + Cl
-
+
e . Similarly, it is known that ionic fragmentation of the
III. Results and Discussion
isovalent nitric acid also yields above the first ionization energy
+
24
(
E g 11.90 eV) the product NO2 .
1
. Fragmentation of the Nitryl Chloride Cation. Figure 1
+
shows a TOF mass spectrum of nitryl chloride, which was
obtained from photoionization with undispersed LPP radiation.
The thermochemical threshold of NO2 formation from
+
ClNO2 is calculated to be 11.06 eV, where the following data
have been used: ∆Hf° (ClNO2) ) 12.134 kJ/mol; ∆Hf° (Cl) )
121.302 kJ/mol; ∆Hf° (NO2) ) 33.095 kJ/mol; IE(NO2) ) 9.586
+
The spectrum shows mass signals corresponding to NO (t )
3
5,37
+
3
3
.59 µs, m/z ) 30),
Cl (t ) 3.88 µs, t ) 3.98 µs; m/z )
+
27,28
+
5, 37), and NO2 (t ) 4.44 µs, m/z ) 46), where the latter
cation is the dominant species in the mass spectrum. The parent
ClNO2 (m/z ) 81, 83) is calculated to occur at t )
.88 µs. However, it is not observed, indicating that it is unstable
eV.
This indicates that the formation of NO2 occurs at the
ionization threshold of ClNO2 with an excess energy of 630 (
3
5,37
+
cation
5
50 meV. The occurrence of excess energy is quite expected,
+
because NO2 formation from ClNO2 requires at least its
with respect to ionic fragmentation. This result is similar to the
isovalent nitric acid (HONO2), which also shows no parent
cation in photoionization mass spectra.24 In addition to the
signals that are related to photoionization of ClNO2, we observe
ionization energy, if ion-pair formation is neglected (see below).
+
Earlier work on HONO2 has shown that the onset of NO2
formation coincides with the thermodynamic threshold, whereas
+
an excess energy of 200 meV has been deduced for NO2
formation from ClONO2.
2
4
weak cation signals that are due to molecular chlorine impurities
+
The photoion yield of NO2+ shows an inflection in the
threshold regime near 12.25 ( 0.05 eV, i.e., ≈560 meV above
the ionization threshold. This feature is expected to be related
(
Cl2 , m/z ) 70, 72, 74; cf. Experimental Section).
+
+
+
Figure 2 shows the photoion yields of NO2 , NO , and Cl .
These cations are formed as a result of fragmentation of ClNO2 .
The onset of the NO2 yield is found at E ) 11.69 ( 0.05 eV.
+
+
+ 2
to the formation of excited ClNO2 ( A2), corresponding to the
removal of an electron from the 1a2 orbital. Frost et al. have
This energy is somewhat lower (∆E ) 150 meV) than the
adiabatic ionization energy IEad that has been obtained earlier
from He(I) photoelectron spectroscopy (IEad(ClNO2) ) 11.84
2
-1
estimated that the adiabatic ionization energy IE( A2) (1a2
)
occurs at 12.40 eV, i.e., 560 meV above the adiabatic ionization
18
18
eV ). We note that somewhat lower and likely more reliable
adiabatic ionization energies are obtained from photoionization
threshold, which is in excellent agreement with the present
results. Another inflection is found near 13.0 eV and is
25
2
mass spectrometry compared to other experimental approaches.
tentatively assigned to the formation of the excited ( A1) state
+
Therefore, we believe that the present value is of superior quality
compared to previous results.18 The appearance of NO2 right
at the photoionization threshold indicates that the ground state
of ClNO2 , which is in accordance with previous results from
+
18
+
photoelectron spectroscopy. The NO2 yield shows a broad
maximum around 14 eV that is most likely due to autoionization.
We note that an unequivocal assignment of this feature cannot
be given, because a broad feature leaves some ambiguity,
especially since the VUV absorption cross section of ClNO2
has not been published to the best of our knowledge. The
photoelectron spectrum shows a wide Franck-Condon gap
+
2
of ClNO2 ( B2) is unstable with respect to fragmentation of
+
the Cl-N bond. NO2 is expected to be the cation that is formed
+
in this process rather than Cl , because the latter cation has the
higher ionization energy, as anticipated from the rule of
Stevenson and Audier.26 As a result, the following process

Photofragmentation of Nitryl Chloride
J. Phys. Chem. A, Vol. 105, No. 20, 2001 4847
between 14 and 18 eV.18 There is a broad photoelectron band
between 18.1 and 19.6 eV, which has been assigned to the
removal of electrons from the 1b1, 4a1, and 2b2 orbitals,
respectively. Therefore, it is expected that there cannot be any
distinct autoionization features in the photoion yields of nitryl
chloride and its fragments in the 14-18 eV regime, which is
consistent with the experimental results shown in Figure 2.
The onset of NO+ formation is observed at E ) 13.93 (
0
+
.07 eV (cf. Figure 2). A plausible fragmentation route is ClNO2
+
-
hν f Cl + NO + O + e . The threshold energy of this
process is calculated to be E ) 13.91 eV by using the reference
2
7
data given above, including ∆Hf° (O) ) 249.18 kJ/mol; ∆Hf°
27
29
(
NO) ) 90.29114 kJ/mol, and IE (NO) ) 9.26438 eV. This
+
indicates that NO is formed without any appreciable amounts
of excess energy at the appearance energy.
The threshold of Cl+ formation is observed at 16.0 ( 0.1
eV. The calculated thermodynamic threshold of this frag-
mentation route is E ) 14.44 eV, corresponding to the fol-
+
-
lowing process: ClNO2 + hν f Cl + NO2 + e , with IE(Cl)
3
0
)
12.96764 eV. An additional smooth onset is found at
E ) 17.7 ( 0.1 eV. It corresponds to the thermodynamic
threshold of 17.61 eV of the process ClNO2 + hν f Cl
+
+
-
NO + O + e .
We finally note that possible decay processes that lead to
3
1
ion-pair formation are expected to be of minor importance,
so that they give negligible contributions to the cation yields
shown in Figure 2. For example, the threshold of the process
ClNO2 + hν f Cl- + NO2 is calculated to be at 7.44 eV,
i.e., far below the first ionization energy. The process ClNO2
+
+
-
30
+
hν f Cl + NO2 is expected to occur above 12.16 eV.
Consistently, we cannot find any evidence for ion pair formation
in the photoion yields shown in Figure 2.
Figure 3. Photolysis mass spectra of nitryl chloride obtained from
various UV-photolysis and VUV-photoionization conditions. (a) Pri-
mary UV photolysis: λ ) 240 nm; subsequent photoionization: E )
The results from photoionization mass spectrometry are of
central importance with respect to the reliable determination of
photolysis quantum yields of electronically excited neutral
ClNO2, as will be outlined in the following section. We note
1
4.10 eV. (b) Primary UV photolysis: λ ) 240 nm; subsequent
photoionization: E ) 13.45 eV. (c) Primary UV photolysis: λ ) 240
nm, subsequent photoionization: E ) 15.01 eV. (d) Primary UV
photolysis: λ ) 308 nm, subsequent photoionization: E ) 14.10 eV.
+
+
that formation of neither Cl nor O is expected to play a role
in photoionization of ClNO2 at photon energies below 16 eV.
However, both atomic cations are formed efficiently by pho-
toionization from the corresponding atomic neutrals, which are
formed upon UV photolysis.6
signals differ considerably between the mass spectra that are
shown in Figure 3. This is essentially due to variations in the
photolysis wavelength as well as the photon energy of the
ionizing VUV-radiation. The intensities of the mass signals are
used in the following in order to determine (i) the quantum state
of the primary atomic photoproducts and (ii) the primary
quantum yields of the competing UV photolysis processes.
The photolysis mass spectra shown in Figures 3a-c have
been recorded at constant photolysis wavelength (λ ) 240 nm).
All photofragmentation routes that are shown in eqs 1a-d are
energetically accessible at this wavelength. The VUV photon
energy was selectively set to 13.45, 14.10, and 15.01 eV,
respectively. These photon energies were chosen in order to
obtain quantitative information on photoionization of the atomic
photolysis products, where we used primarily sensitive cation
detection via autoionization, as a probe of the corresponding
quantum states. The variation of the VUV photon energy is also
used to exclude ionic fragmentation channels of ClNO2, as
discussed in the previous section.
2
. Primary Photolysis Products of Electronically-Excited
Nitryl Chloride. The UV photolysis of nitryl chloride has been
investigated by using the pump-probe detection scheme, as
outlined in the Experimental Section. Figure 3 shows a series
of photolysis mass spectra that have been obtained from different
excitation conditions. The spectra shown in Figure 3 (a-c) have
been recorded at the primary UV photolysis wavelength λ )
2
40 nm, whereas λ ) 308 nm was used in the case of the
spectrum that is displayed in Figure 3d. Various mass signals
are observed that are assigned to the occurrence of O , NO ,
and
+
+
3
5,37
+
+
Cl . In addition, NO2 is found to be present in all
photolysis spectra. This mass line is not shown in Figure 3,
because it is also found without primary UV photolysis.
Therefore, we cannot assign this cation to any photolysis channel
of electronically excited ClNO2, rather than to dominant
contributions from ionic fragmentation routes (see previous
The photolysis mass spectra shown in Figure 3 give clear
+
+
+
section). In contrast, the O and Cl mass signals are entirely
evidence for an intense Cl signal. This mass signal is due to
absent in photoionization mass spectra that are recorded at E
photoionization of Cl that is formed according to eq 1a or 1c at
photon energies that are greater than the first ionization energy
<
16 eV, if there is no primary UV photolysis. The exclusive
3
0
occurrence of these atomic cations in pump-probe mass spectra
is a clear indication for the formation of the corresponding
neutrals upon UV photolysis, as anticipated from eqs 1(a-d).
However, it is found that the relative intensities of the mass
of atomic chlorine (IE(Cl) ) 12.96764 eV ). Other processes
+
that may contribute to Cl formation are discounted, such as
+
(i) fragmentation of ClNO2 , because the threshold energy is
found to be 16.0 ( 0.1 eV (see previous section); (ii) ionic

4848 J. Phys. Chem. A, Vol. 105, No. 20, 2001
Plenge et al.
fragmentation of photolytically formed ClNO, because the
(iV) Ionic fragmentation of NO2 that is produced Via eq 1a.
+
+
-
+
+
threshold of the process ClNO + hν f Cl + NO + e is
calculated to occur at E ) 16.628 eV;32 (iii) photolysis of Cl2
impurities in the sample (cf. Figure 1), since the absorption cross
section of Cl2 at the photolysis wavelength λ ) 240 nm is low
The appearance energy of NO formation from NO2 is 12.34
41
eV.
We conclude that the NO+ signal can have, in principle,
contributions from all processes discussed above. The impor-
tance of route (i) cannot be evaluated on the basis of the present
-
22
2 33
(
(
σ(Cl2) ) 4 × 10
ClNO2) ) 1.4 × 10
Assuming that atomic chlorine (Cl( P)) is the dominant source
cm ) compared to that of ClNO2 (σ-
-
18
2 5,6
+
cm ).
results. It is likely that the dominant contribution to the NO
2
signal comes from routes (ii) and (iv), especially because NO2
is formed efficiently in excited states by photolysis ClNO2.7
This point is discussed in more detail in the following section.
3. Determination of Primary UV-Photolysis Quantum
Yields. UV photolysis of nitryl chloride at λ ) 240 nm allows,
in principle, the competitive and simultaneous formation of
atomic photolysis products, such as Cl and O (cf. eq 1). In this
-9
+
6,9
of the Cl mass signal, one can derive a common scale of
the vertical axis of the photolysis mass spectra shown in Figures
a-c. This is accomplished by scaling the intensity of the Cl+
3
signal proportional to the photoionization cross section of
2
34
Cl( P).
The photolysis mass spectra shown in Figures 3a and 3b have
been recorded at 14.10 and 13.45 eV, respectively. Experiments
at both photon energies allow us to obtain specific information
on the quantum state of atomic oxygen that is formed upon
ClNO2 photolysis. Photoionization and autoionization of ground-
+
+
case, the analysis of the Cl and O intensities cannot be used
to obtain absolute quantum yields for the processes correspond-
ing to eqs 1a-c so that rather the particle ratio between the
nascent atoms O and Cl is determined, which is of primary
importance to atmospheric implications (cf. section 4).
3
state O ( P) has been investigated in detail previously, where
+
+
The intensity ratio of the ionized photoproducts O and Cl
I(O )/I(Cl )) is given by
various autoionization features were observed in the VUV
+
+
(
regime.3
5-37
2
2
4
3
I(O+)
I(Cl+)
The autoionizing multiplet transition 1s 2s 2p ( P) f
γ(1b) + γ(1c) σ(O)
γ(1a) + γ(1c) σ(Cl)
2
2
3
3
3
)
×
(2)
1
s 2s 2p 3s′′ ( P) occurs near E ) 14.10 eV, so that O ( P) can
be selectively probed at this photon energy (cf. Figure 3a). The
ionization energy of O ( P) is IE ) 13.61806 eV. Thus, O
P) cannot be photoionized by 13.45 eV photons. Photoexci-
3
30
with
3
(
tation at E ) 13.45 eV leads exclusively to the selective
γ(1b) + γ(1c) N(O)
)
1
4
3
(3)
detection of O ( D), since the 2p f 2p d′ transitions occur
close to this photon energy. These are known to occur as a
single, intense autoionization resonance at low spectral resolu-
γ(1a) + γ(1c) N(Cl)
where the γ values correspond to the relative yields of the
processes of eq 1, σ is the photoionization cross section, and
N(O)/N(Cl) is the nascent branching ratio of atomic oxygen
and atomic chlorine.
tion.16 The photolysis mass spectra displayed in Figures 3a and
+
3
b give clear evidence that O is formed exclusively upon
3
photoionization of O ( P), whereas we cannot find any evidence
for photoionization of O ( D) upon photolysis of ClNO2 at λ )
1
The photoionization cross section of the ground-state atoms
3
2
2
40 nm (cf. eq 1d).
O ( P) and Cl ( P) has been reported earlier (cf. refs 34,35,-
42-46). It is reliably determined in spectral regimes where
exclusively transitions into photoionization continua occur, so
that the bandwidth of the VUV source has no influence on the
cation signal strength. This is unlike narrow autoionization
resonances that are superimposed to the ionization continua.
For example, it is well known that the relative photoionization
+
NO is observed in the photolysis spectra (cf. Figure 3). This
cation signal may have various origins:
i) Photoionization of neutral, nascent NO that is formed
(
according to eq 1c. Additional experiments have been performed
+
in order to obtain further information on the origin of the NO
mass signal. Time-of-flight mass spectra have been recorded at
3
4
cross section of O ( P) remains almost constant in the S°
E ) 13.82 eV, where NO is known to autoionize as a result of
4
continuum, i.e., between the S° ionization threshold (IE )
2
3
38
the X( Π, V ) 0) f 3pπ b( Π) Rydberg transition. These mass
3
0
1
3.61806 eV ) and 14.10 eV, where the onset of intense
+
spectra did not show any evidence for an NO yield that is
3
5
autoionization features is observed.
resonantly enhanced via autoionization. This indicates that the
We have chosen for photoionization experiments the photon
energy E ) 15.01 eV. This photon energy allows us to determine
reliably the branching ratio between both atomic products (cf.
+
NO mass signal that is shown in Figure 3 cannot be a result
of photoionization of neutral ground state NO. However, if
nascent NO is exclusively formed in its vibrationally excited
3
Figure 3c), because the photoionization cross sections of O ( P)
2
ground state upon UV photolysis of ClNO2, the X( Π, V > 0)
2
3
42
and Cl ( P) are well known: σ(O ( P)) ) 3.3 Mb and σ(Cl
3
f 3pπ b( Π) Rydberg transition does not occur at 13.82 eV so
2
34
(
P)) ) 30.5 Mb. Moreover, the latter value is in full agreement
that process 1c may not be probed efficiently. Thus, we cannot
4
7
with theoretical work. The relative error limits of the
10
discount this channel, which is in accordance with earlier work.
42,34
experimental values are <9% and ∼20%, respectively.
(
ii) Fragmentation of excited NO2 that is formed Via reaction
+
+
The experimental intensity ratio of the Cl /O signal is found
1
(a), yields NO+O. This channel can occur only if NO2 is
formed upon the photolysis of ClNO2 in an excited state.
+
+
to be I(Cl )/I(O ) ) 13.10 ( 1.79 (cf. Figure 3c). The sum of
the quantum yields of the processes 1a-c is given by γ(1a) +
γ(1b) + γ(1c) ) 1. This assumption is valid, because the
photolytical formation of products that contribute to the channel
Previous photolysis experiments on NO2 have shown that λ <
2
3
39
4
30 nm yields the products NO ( Π) + O ( P). Evidence for
the formation of excited NO2 comes from recent photolysis
1
d is ruled out (see above). This also implies that other possible
experiments of ClNO2 at 235 and 248 nm.7
-9
decay processes of UV-excited ClNO2 that may compete with
the above-mentioned photochemical decay routes, such as ion
pair formation or radiative relaxation, are of minor importance.
We note that the primary quantum yields of the processes 1a-c
cannot be obtained without ambiguity, because we detect atomic
(iii) Ionic fragmentation of ClNO that is formed according
to eq 1(b). The first ionization energy of ClNO is IE ) 10.90
3
0
+
(
0.03 eV. The parent cation ClNO is known to be unstable
+
40
with respect to NO formation.

Photofragmentation of Nitryl Chloride
J. Phys. Chem. A, Vol. 105, No. 20, 2001 4849
products that can occur in different photolysis routes at λ )
(3P) formation is of minor importance at λ > 300 nm, but the
+
+
2
40 nm. Consequently, the Cl /O intensity ratio that is obtained
yield of this channel is enhanced at shorter wavelengths, which
10
from photoionization mass spectrometry can be used to deter-
mine the ratio for the formation of atomic products, yielding at
λ ) 240 nm: N(Cl)/N(O) ) 1.44 ( 0.15.
is in agreement with earlier work. A comparison of the present
results with previous investigations shows that the quantum yield
for Cl formation at λ ) 350 nm is identical to that reported in
the present work at λ ) 308 nm (cf. ref 6). However, we note
that (i) the error limits are reduced compared to previous work,6
(ii) we have obtained a specific value for the quantum yield of
ClNO + O formation rather than an upper limit (cf. ref 6), (iii)
the error limit of γ does not depend in the present approach on
the uncertainty of the UV absorption cross section, and (iv) the
current approach has the inherent advantage that both atomic
photolysis products are detected simultaneously and independent
of the kinetic energy of the photofragments.
We note that in recent work no atomic oxygen was observed
7
at all upon photolysis at λ ) 248 nm, whereas in earlier work
1
0
a contribution of 20% was ascribed to channel 1c. This
discrepancy in previous results cannot be resolved. We assign
3
the comparably high intensity of O ( P) that is detected in the
present work via photoionization with high sensitivity and
selectivity primarily to processes 1a and 1c. In addition, there
are contributions to channel 1c, which is believed to occur with
small quantum yield at λ ) 248 nm.7
,10,11
The present results
are rationalized as follows: ClNO2 decays with a yield of ∼90%
via reactions 1a and 1c, where we cannot distinguish between
the concerted and sequential decay into Cl + NO + O. On the
other hand, channel 1b yields atomic oxygen with a yield of
4. Atmospheric Implications. It has been shown in the
previous section that the quantum yield of atomic chlorine
formation as a result of ClNO2 photolysis remains unchanged
between λ ) 308 nm and λ ) 350 nm (cf.ref 6). From this it
is expected that the quantum yields of Cl and O formation from
ClNO2 remain almost constant in the UV regime of solar
radiation that penetrates the lower atmosphere. Photolysis is the
dominant decomposition pathway of atmospheric ClNO2, be-
cause competing chemical processes, such as the reaction with
∼
10%, which is in agreement with present results that are
obtained from the photolysis at λ ) 308 nm (see below).
Furthermore, we consider that ∼40% of the excited NO2 will
dissociate into NO + O. Finally, ∼20% of the photolysis of
ClNO2 populates channel 1c, according to ref 10. These
assumptions yield N(Cl)/N(O) ) 1.41, indicating that other
4
8
OH, are known to be comparably slow. This implies that
∼93% of the tropospheric ClNO2 yields atomic chlorine upon
photolysis, where NO2 is the companion photoproduct (cf. eq
3
sources of O ( P) formation can be discounted, such as
subsequent photoabsorption of nascent NO2 by the same
photolysis pulse and multiphoton processes, as anticipated from
the low fluence of the photolysis laser.
3
1a). Formation of atomic oxygen O ( P) and the companion
photoproduct ClNO occurs only as a minor channel (cf. eq 1b).
49
Additional photolysis experiments have been carried out at
λ ) 308 nm, which is already in the regime of the actinic flux
that penetrates the lower atmosphere. Reaction 1c cannot occur
at this photolysis wavelengths, i.e., γ(1c) ) 0 and γ(1d) ) 0,
i.e., γ(1a) + γ(1b) ) 1. Thus, absolute values of the primary
quantum yield can be derived without any ambiguity from the
branching ratio of the atomic products. The branching ratio
N(O)/N(Cl) is then identical to the ratio of the quantum yields
of the reactions 1a and 1b (cf. eqs 2-3) yielding
There are three decay routes of atmospheric ClNO: (i) reaction
with OH, (ii) hydrolysis, and (iii) absorption of UV radiation
and subsequent photolytic decay. The absorption cross section
-
19
of ClNO shows a weak maximum with σClNO ) 1.52 × 10
2
50
cm at λ ) 340 nm. Photoexcitation leads to photolysis
according to ClNO + hν f Cl + NO with a quantum yield of
unity. It has been shown that the photolytic lifetime of ClNO
is 5 min at a zenith angle of 0°, and 27 min at 80°. As a
result, ClNO2 will yield Cl + NO + O after its primary
photolysis, according to eq 1b, and subsequent photolysis of
the primary fragment ClNO. Photolytical formation of Cl +
NO2 (cf. eq 1a) is expected to be followed by the rapid
5
1
49
+
γ(1b) I(O )σ(Cl)
)
(4)
+
γ(1a)
52
I(Cl )σ(O)
photolysis of NO2 during daytime. As a result, UV photolysis
of ClNO2 yields the same final products no matter if reactions
Photoionization at 15.01 eV yields no detectable O+ signal,
indicating that reaction 1b is of minor importance at λ ) 308
nm. This result is in full agreement with earlier results, where
1
a or 1b occur. Thus, ClNO2 is a photolytical source of atomic
chlorine, which competes with OH in tropospheric photooxi-
dation cycles and is expected to add atomic chlorine to the
stratosphere as a result of its heterogeneous formation on
stratospheric clouds or aerosols after the polar night.
Finally, we note that the present results are obtained under
collision-free conditions in the laboratory, where the primary
photoproducts are probed ≈100 ns after the primary photolysis.
However, in the atmosphere collisional deactivation or long-
lived electronic states of nascent photoproducts may occur. Both
processes will not change the formation of the final products
Cl + NO + O, which subsequently undergo chemical stabiliza-
tion in the atmosphere.
6
λ ) 350 nm was used for photolyzing nitryl chloride. The
+
sensitivity for the detection of O is significantly enhanced if
the VUV photon energy is set to 14.10 eV, because at this
3
4
3
3
photon energy O ( P) is autoionizing via the 2p ( P) f 2p 3s′′
3
+
(
P) transition (see above). A weak O signal is detected under
these conditions (cf. Figure 3d).The quantum yield of Cl ( P)
and O ( P) formation at λ ) 308 nm is reliably deduced from
the photolysis mass spectrum shown in Figure 3d, because the
O /Cl intensity ratio between on-resonance (14.10 eV, cf.
Figure 3a) and off-resonance (15.01 eV, cf. Figure 3c) measure-
ments is established from experiments at λ ) 240 nm, where
the atomic oxygen occurs exclusively in the P-ground state
see above). From Figure 3a one obtains I(Cl )/I(O ) ) 2.50
0.09, whereas at λ ) 308 nm the intensity ratio is significantly
enhanced yielding I(Cl )/I(O ) ) 22.97 ( 3.89. This corre-
sponds to the primary quantum yields at λ ) 308 nm: γ308
nm(Cl + NO2) ) 0.93 ( 0.10 and γ308 nm (ClNO + O) ) 0.07
2
3
+
+
3
IV. Conclusions
+
+
(
(
The UV photolysis of nitryl chloride (ClNO ) has been
2
+
+
investigated at λ ) 240 nm and λ ) 308 nm. Photoionization
mass spectrometry is shown to be a suitable detection method
for obtaining quantitative information on the photolysis products,
where tunable VUV radiation was generated in a laser-produced
plasma. The experimental approach allowed us to determine
unequivocally the quantum state of atomic oxygen that is formed
(
0.01.
2
These results indicate that Cl ( P) formation is the dominant
photolysis pathway at λ g 240 nm, similar to recent work. O
7

4850 J. Phys. Chem. A, Vol. 105, No. 20, 2001
Plenge et al.
3
upon photolysis, where exclusively O ( P) is detected. Moreover,
(20) Schott, G.; Davidson, N. J. Am. Chem. Soc. 1958, 80, 1841.
(21) Volpe, M.; Johnston, H. S. J. Am. Chem. Soc. 1956, 78, 3903.
22) Bernitt, D. L.; Miller, R. H.; Hisatsune, I. C. Spectrochim. Acta
967, 23A, 237.
23) Orphal, J. private communication, 2000.
(24) Jochims, H. W.; Denzer, W.; Baumg a¨ rtel, H.; L o¨ sking, O.; Willner,
H. Ber. Bunsen-Ges. Phys. Chem. 1992, 96, 573.
25) R u¨ hl, E.; Rockland, U.; Baumg a¨ rtel, H.; L o¨ sking, O.; Binnewies,
M.; Willner, H. Int. J. Mass Spectrom. 1999, 185/186/187, 545.
26) Stevenson, D. P. Discuss. Faraday Soc. 1951, 10, 35. Audier, H.
3
the branching ratio of the primary atomic photoproducts O ( P)
and Cl ( P) changes with photolysis wavelength, where the latter
product dominates the long wavelength regime (λ > 308 nm).
The atmospheric fate of the primary products is briefly
discussed. The use of photoionization mass spectrometry also
allowed us to investigate the photoion yields of the cations that
(
2
1
(
(
+
are formed upon photoionization from ClNO2 .
(
E. Org. Mass Spectrom. 1969, 2, 283.
Acknowledgment. We thank Dr. J. Orphal for helpful
discussions on the synthesis of nitryl chloride and for providing
the infrared absorption spectrum of ClNO2. J.P. acknowledges
financial support by the Deutsche Bundesstiftung Umwelt.
Financial support by the Bundesministerium f u¨ r Forschung und
Bildung (BMBF) (grant no.: 01 LO 9610/3) and by the Fonds
der Chemischen Industrie are gratefully acknowledged.
(27) Chase, M. W., Jr., NIST-JANAF Thermochemical Tables, 4th
Edition J. Phys. Chem. Ref. Data 1998, Monograph 9, 1.
(
28) Haber, K. S.; Zwanziger, J. W.; Campos, F. X.; Wiedmann, R. T.;
Grant, E. R. Chem. Phys. Lett. 1988, 144, 58.
29) Reiser, G.; Habenicht, W.; M u¨ ller-Dethlefs, K.; Schlag, E. W.
Chem. Phys. Lett. 1988, 152, 119.
30) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
(
(
R. D.; Mallard, W. G. Ion Energetics Data in NIST Chemistry Webbook,
NIST Standard Reference Database Number 69; Mallard, W. G, Linstrom,
P. J., Eds.; National Institute of Standards and Technology: Gaithersburg
MD, 1998.
References and Notes
(
31) Rockland, U.; Baumg a¨ rtel, H.; R u¨ hl, E.; L o¨ sking, O.; M u¨ ller, H.
S. P.; Willner, H. Ber. Bunsen-Ges. Phys. Chem. 1995, 99, 969.
32) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
R. D.; Mallard, W. G. J. Phys. Chem. Ref. Data 1988, 17, Suppl. 1.
(
1) Behnke, W.; George, C.; Scheer, V.; Zetzsch, C. J. Geophys. Res.
1
997, 102, 3795.
(
(
(
2) Behnke, W.; Scheer, V.; Zetzsch, C. J. Aerosol Sci. 1993, 24, 115.
3) Tolbert, M. A.; Rossi, M. J.; Golden, D. M. Science 1988, 240,
(
(
(
33) Hubinger, S.; Nee, J. B. J. Photochem. Photobiol. 1995, 86, 1.
34) Rus cˇ ic, B.; Berkowitz, J. Phys. ReV. Lett. 1983, 50, 675.
35) Dehmer, P. M.; Berkowitz, J.; Chupka, W. A. J. Chem. Phys. 1973,
1
6
018.
4) Finlayson-Pitts, B. J.; Livingston, F.; Berko, H. Nature 1990, 343,
(
22.
5
6
9, 5777.
(
(
(
5) Illies, A. J.; Takacs, G. A. J. Photochem. 1976/77, 6, 35.
6) Nelson, H. H.; Johnston, H. S. J. Phys. Chem. 1981, 85, 3891.
7) Furlan, A.; Haeberli, M. A.; Huber, J. R. J. Phys. Chem. A 2000,
(
(
36) Dehmer, P. M.; Chupka, W. A. J. Chem. Phys. 1975, 62, 584.
37) Dehmer, P. M.; Luken, W. L.; Chupka, W. A. J. Chem. Phys. 1977,
7, 195.
1
3
(38) Erman, P.; Karawajczyk, A.; Rachlew-K a¨ llne, E.; Str o¨ mholm, C.
J. Chem. Phys. 1995, 102, 3064.
(39) Roehl, C. M.; Orlando, J. J.; Tyndall, G. S.; Shetter, R. E.; Vasquez,
G. J.; Cantrell, C. A.; Calvert, J. G. J. Phys. Chem. 1994, 98, 7837.
(40) Syrvatka, B. G.; Gil’burd, M. M. Russ. J. Phys. Chem. 1973, 47,
1215.
(41) Dibeler, V. H.; Walker, J. A. AdV. Mass Spectrom. 1967, 4, 767.
(42) Angel, G. C.; Samson, J. A. R. Phys. ReV. A 1988, 38, 5578.
(
07.
12) Farman, J. C.; Gardiner, B. G.; Shanklin, J. D. Nature 1985, 315,
2
(43) Comes, F. J.; Speier, F.; Elzer, A. Z. Naturforsch. 1968, 23a, 125.
(44) Cairns, R. B.; Samson, J. A. R. Phys. ReV. 1965, 139A, 1403.
(45) Kohl, J. L.; Lafyatis, G. P.; Palenius, H. P.; Parkinson, W. H. Phys.
(
13) Flesch, R.; Wassermann, B.; Rothmund, B.; R u¨ hl, E. J. Phys. Chem.
1
994, 98, 6263.
(
14) Flesch, R.; Sch u¨ rmann, M. C.; Hunnekuhl, M.; Meiss, H.; Plenge,
J.; R u¨ hl, E. ReV. Sci. Instrum. 2000, 71, 1319.
15) Flesch, R.; Sch u¨ rmann, M. C.; Plenge, J.; Hunnekuhl, M.; Meiss,
H.; R u¨ hl, E. Phys. Chem. Chem. Phys. 1999, 1, 5423.
16) Flesch, R.; Sch u¨ rmann, M. C.; Plenge, J.; Meiss, H.; Hunnekuhl,
M.; R u¨ hl, E. Phys. ReV. A 2000, 62, 52723.
17) Pallix, J. B.; Sch u¨ hle, U.; Becker, C. H.; Huestis, D. L. Anal. Chem.
989, 61, 805.
18) Frost, D. C.; Lee, S. T.; McDowell, C. A.; Westwood, N. P. C. J.
Electron Spectrosc. Relat. Phenom. 1975, 7, 331.
19) Samson, J. A. R. Techniques of Vacuum UltraViolet Spectroscopy;
Wiley: New York, 1967.
ReV. A 1978, 18, 571.
(
(
(
46) Berkowitz, J. J. Phys. B. 1997, 30, 583.
47) Tayal, S. S. Phys. ReV. A 1993, 47, 182.
48) Ganske, J. A.; Ezell, M. J.; Berko, H. N.; Finlayson-Pitts, B. J.
(
(
Chem. Phys. Lett. 1991, 179, 204.
49) Finlayson-Pitts, B. J. Nature 1983, 306, 676.
(
(
1
(50) Tyndall, G. S., Stedman, K. M.; Schneider, W.; Burrows, J. P.;
Moortgat, G. K. J. Photochem. 1987, 36, 133.
(51) Grimley, A. J.; Houston, P. L. J. Chem. Phys. 1980, 72, 1471.
(52) Calvert, J. G.; Pitts, J. N., Jr. Photochemistry; Wiley: New York,
1966.
(
(