State-resolved photofragmentation of [ClNO]n van der Waals clusters in a supersonic jet

Article abstract of DOI:10.1021/jp9840127

The effects of the ultraviolet laser irradiation of [ClNO]₂, weakly bound clusters, formed in a supersonic jet, are analyzed by considering three processes: the photofragmentation of bare ClNO, the Cl+ClNO reaction, and NO relaxation within the cluster. The photofragmentation of jet-cooled ClNO at 355 nm produces NO (υ″ = 1) with a kinetic energy of 2240 cm^-1, a spin-orbit preference of F₁/F₂ = 1.2, and Λ-doublet state preferences of Π(A″)/Π±(A″) = 2.0 and 4.0 for the F₁ and F₂ manifolds, respectively. The NO distribution of rotational states was parametrized using a Gaussian function centered at N = 34, with a fwhm of 17. On the other hand, the Cl+ClNO reaction, studied at a collision energy of 2780 cm^-1, gives NO(υ″ = 1) described by a Boltzmann rotational distribution with T_rot = 950±100 K. The relative population of the NO spin-orbit states is F₁/F₂ = 2.5, with a Λ-doublet state preference of Π(A″)/Π(A′) = 1.2 and E_trans(NO) of 578 cm^-1. It is found that 57% of available energy is disposed as E_int(Cl₂). As a result of the irradiation of the [ClNO]_n, clusters at 355 nm are observed: Boltzmann ensembles of NO(υ″ = 1) and NO(υ″ = 0) molecules described by T_rot of 310±30 and 170±25 K, respectively, with no spin-orbit or Λ-doublet state preferences, overlapped with a Gaussian distribution already assigned to the NO photofragment. The relative contribution of the NO(υ″ = 1) photofragment to the spectra is drastically reduced upon increasing the backing pressure, as it undergoes translational and rotational relaxation within the clusters. Our high-resolution studies provide evidence that suggests that the reaction takes place within the [ClNO]_n clusters.

Full text of DOI:10.1021/jp9840127

J. Phys. Chem. A 1999, 103, 1929-1938
1929
ARTICLES
State-Resolved Photofragmentation of [ClNO]n van der Waals Clusters in a Supersonic Jet
Carlos Conde,† Christof Maul,‡,§ and Edwin Quin˜ones*
Department of Chemistry, UniVersity of Puerto Rico, P.O. Box 23346, San Juan, Puerto Rico 00931-3346
ReceiVed: October 8, 1998; In Final Form: February 3, 1999
The effects of the ultraviolet laser irradiation of [ClNO]n weakly bound clusters, formed in a supersonic jet,
are analyzed by considering three processes: the photofragmentation of bare ClNO, the Cl + ClNO reaction,
and NO relaxation within the cluster. The photofragmentation of jet-cooled ClNO at 355 nm produces NO-
(υ′′ ) 1) with a kinetic energy of 2240 cm-1, a spin-orbit preference of F1/F2 ) 1.2, and Λ-doublet state
preferences of Π(A′′)/Π(A′′) ) 2.0 and 4.0 for the F1 and F2 manifolds, respectively. The NO distribution
of rotational states was parametrized using a Gaussian function centered at N ) 34, with a fwhm of 17. On
the other hand, the Cl + ClNO reaction, studied at a collision energy of 2780 cm-1, gives NO(υ′′ ) 1)
described by a Boltzmann rotational distribution with Trot ) 950 ( 100 K. The relative population of the NO
spin-orbit states is F1/F2 ) 2.5, with a Λ-doublet state preference of Π(A′′)/Π(A′) ) 1.2 and Etrans(NO) of
578 cm-1. It is found that 57% of available energy is disposed as Eint(Cl2). As a result of the irradiation of
the [ClNO]n, clusters at 355 nm are observed: Boltzmann ensembles of NO(υ′′ ) 1) and NO(υ′′ ) 0) molecules
described by Trot of 310 ( 30 and 170 ( 25 K, respectively, with no spin-orbit or Λ-doublet state preferences,
overlapped with a Gaussian distribution already assigned to the NO photofragment. The relative contribution
of the NO(υ′′ ) 1) photofragment to the spectra is drastically reduced upon increasing the backing pressure,
as it undergoes translational and rotational relaxation within the clusters. Our high-resolution studies provide
evidence that suggests that the reaction takes place within the [ClNO]n clusters.
I. Introduction
they found that OH is produced rotationally colder than when
the same reaction was studied under bulk conditions because
the range of impact parameters available to the reaction is
narrowed as a consequence of the stereochemical control of the
collision.8 On the other hand, Sauder et al. in their study of the
16O(1D) + H218O reaction, photoinitiated within the O3-H2O
vdW complex, determined the rotational distributions of the two
OH products, one of which carries the information of the newly
formed bond.9 They concluded that a significant fraction of the
available energy to the reaction is removed by O2, the other
product of the photolysis of ozone, before departing from the
collision region. Studies of reactions involving vdW molecules
containing metal atoms initiated with polarized laser light have
contributed enormously to the understanding of orbital alignment
effects in chemical reactions as exemplified by the work of
Breckenridge, Soep, Jouvet, and co-workers.10,11 Moreover, vdW
complexes have been the subject of energy transfer studies,
which should be mentioned: the vibrational-to-translational
energy transfer investigation within the HF-NO vdW complex
carried out by Shorter et al.,12 the study of Honma et al.13 of
the atom + diatomic inelastic collisions within the [NO]2 vdW
complex, and predissociation studies of van der Waals
molecules.14-17
Beyond their intrinsic importance, van der Waals complexes
provide us with the prospect of studying chemical processes in
environments with unprecedented physical properties.1,2 While
in the past it was necessary to disregard those factors that do
not have a direct influence on the outcome of elementary
reactions, these weakly bound clusters are now amenable to
explicitly consider the individual interactions between the
reactants and the surrounding solvent molecules during the
trajectory of a collision.3,4 The understanding that has emerged
of the solvation problem together with the tremendous progress
achieved in the past two decades on the characterization of
molecular interactions employing spectroscopic and theoretical
methods, has encouraged this research direction.5-7
Through the appropriate choice of the gas expansion condi-
tions it is possible to favor the formation of binary van der Waals
(vdW) precursors to study photochemical reactions. Solvent
effects do not play a role in this class of experiments, and the
results derived from them aim at improving the understanding
of the dynamics of elementary molecular collisions. In particular,
in the investigation carried out by Wittig and co-workers on
the H + CO2 reaction preparing the BrH-CO2 vdW precursor,
† Present address: SmithKline Beecham Pharmaceuticals Co., P.O. Box
11975, Cidra, Puerto Rico 00639-1975.
Reactions initiated in vdW clusters tend to display the
behavior observed in the liquid state as the cluster size
increases.1 Garvey and Bernstein studied reactions of methyl
halides induced by electron impact ionization in the environment
of weakly bound clusters.18 This was the first experimental
‡ Present address: Institut Fu¨r Physikalische und Theoretische Chemie,
Technische Universita¨t Braunschweig, Hans-Sommer-Strasse 10, D-38106
Braunschweig, Germany.
§ Fellow of the Deutsche Forschungsgemeinschaft.
* Address correspondence to this author.
10.1021/jp9840127 CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/12/1999

1930 J. Phys. Chem. A, Vol. 103, No. 13, 1999
Conde et al.
attempt to test whether the existing kinetic theories that describe
the behavior of ion-molecule reactions could be employed to
explain their behavior within clusters. Sivakumar et al. carried
out a photofragmentation investigation of [OCS]n clusters. In
their work, OCS was the light absorbing species, OCS + hν f
S(1D) + CO, and carbon monoxide was detected rotationally
and vibrationally cool as a consequence of the relaxation process
in the clusters. Sivakumar et al. observed the S(1D) + OCS f
S2 + CO reaction within the [OCS]n clusters by monitoring the
S2 molecule and concluded that this reaction was another source
of CO.19 Bergmann and Huber photodissociated methyl nitrite,
CH3ONO + hν f CH3O + NO, solvated in large [CH3ONO]n
clusters and detected NO from the photodissociation of bare
CH3ONO, NO rotationally relaxed emanating from the clusters,
and the CH3O + NO f HNO + H2CO disproportionation
reaction.20 Jones et al. found that the elimination of SF6 was
the dominant process during the infrared photofragmentation
of SF6[C6H6]n+ clusters.21 Ning and Pfab studied the photodis-
sociation of nitrous acid embedded in water clusters and
determined a bimodal NO rotational distribution corresponding
to the cases when HONO was electronically excited under
collisionless conditions and within the water clusters, respec-
tively.22 On the other hand, Rudich et al. found that cyclohexanol
(the product observed in liquid cyclohexane) is formed when
the O(3P) + C6H10 reaction was carried out in cyclohexane
clusters, while the OH and cyclohexyl radicals are the products
formed under single collision conditions, clearly indicating that
the abstraction mechanism is suppressed by the cluster cage.23
In this paper we present a study of the dynamics of the
photofragmentation of [ClNO]n weakly bound clusters at 355
nm, formed in a supersonic jet. Working under expansion
conditions at which the clusters are formed, we observed nitric
oxide (1) that has undergone translational, rotational, and spin-
orbit state relaxation, (2) that is from the ClNO photofragmen-
tation, and (3) that is from the Cl + ClNO reaction. High-
resolution measurements allowed us to assign the contribution
of the three basic processes to different regions of the NO fine
structure spectra. To understand the photofragmentation of the
[ClNO]n clusters, we carried out studies under bulk conditions
of the dynamics of the photofragmentation of ClNO and of the
Cl + ClNO reaction.
photolysis and the probe laser was controlled by a digital delay
and pulse generator (Stanford Research DG535). The laser-
induced fluorescence (LIF) signal was viewed by a frontal PMT
(Thorn EMI), coupled to a 300 MHz digital oscilloscope
(LeCroy 9450A), integrated by a boxcar (Stanford Research SR
250), fed into a Macintosh IIci computer, and interfaced to a
GW Instrument MacAdios II card, programmed with MPW
FORTRAN. Each data point was averaged for at least 10 laser
pulses, and the spectra were corrected for variations in the dye
laser output energy as a function of wavelength. The dye laser
resolution was adjusted to ensure that each rotational line was
defined by at least five data points. Prior to every experimental
run, the 0-0 band origin of the A2Σ+ r X2Π1/2 transition in
NO was monitored in order to establish low-temperature
conditions. We detected the emission emanating from the
photodissociation of the NO-Ar vdW complex around 224 nm
to optimize the delay between the pulsed valve and the
photolysis laser to ensure cluster formation conditions.24
An intracavity Etalon was installed to scan individual
rotational lines. A resolution of 0.04 cm-1 in the UV region
was confirmed by scanning the rotational lines of jet-cooled
NO.
ClNO was prepared by mixing 3:1 samples of NO and Cl2
(Air Products) in the presence of room light, allowing 20 min
for the mixture to reach equilibrium. The excess NO was
pumped out after cooling the sample using a dry ice/methanol
slush. Typical samples consisted of 3% ClNO in He at total
pressures ranging from 1 to 6 bar.
The experiments in the bulk were done in the molecular beam
chamber by shutting off the gate valve and adding 150 mTorr
of ClNO.
B. Spectroscopic Simulations. The experimental distributions
of NO(υ′′ ) 0,1) rotational states were derived using computer
simulations of the NO[A2Σ+(υ′ ) 0) r X2ΠΩ(υ′′ ) 1)]
transitions. We showed for υ′′ ) 0 that the integration of the
lines and the simulations give the same results. We included
75 rotational states in our simulations. The following expression
was employed to calculate the rotational energy of the A2Σ+
state,
ν˜A ) N(N + 1)B˜0′ - N2(N + 1)2D˜ e′
(1)
where N stands for the rotational quantum number excluding
II. Experimental Section
the electron spin according to Hund’s case (b), B˜′ is the
o
rotational constant in υ′ ) 0, and D˜ ′e is the centrifugal
distortion constant.25 The rotational energy of the X2Π1/2 spin
state (F1 manifold) is given by25
A. Description of the Apparatus. Our apparatus consists
of a 12 in diameter stainless steel chamber coupled to a Varian
VHS-10 diffusion pump, pumped by a 39 cfm mechanical pump.
A pulsed solenoid valve (General Valve Series 9) with a 0.8
mm nozzle diameter was used to expand the ClNO/He mixtures
and the reactions were observed at 25 nozzle diameters
downstream. The pressure in the chamber was about 1 × 10-6
Torr before operating the pulsed valve and never was above 1
× 10-4 Torr when the valve was operated at 10 Hz at backing
pressures up to 6 atm. A Nd:YAG laser (Continuum Surelite
II) in its third harmonic (355 nm) was utilized to photodissociate
ClNO using laser pulses of 3 mJ. The photolysis laser beam
was focused onto an area of 1 mm2 using a 100 cm focal length
lens. The dye laser (Lambda Physik LPD 3002) was pumped
by an excimer laser (Lambda Physik LPX 205i) operating at
308 nm. The dye was Coumarin 460 (Aldrich), and the beam
was frequency doubled by a BBOI doubling crystal. The dye
laser beam was collinearly counter propagated to the photolysis
laser to probe the fine structure of NO. The NO molecules were
detected by using probe laser pulses of less than 5 µJ in order
to avoid optical saturation. A time delay of 50 ns between the
ν˜(J′′)F
)
1
1 2
2
1
1 2
2
J′′ +
- 1 -
4 J′′ +
+ Y(Y - 4) B˜′′- J′′ 4D˜ ′′
[
(
]
1
e
)
(
)
2x
(2)
1
where J ) N + /2, A1 is the spin-orbit constant, B˜1′′ is the
rotational constant in υ′′ ) 1, D˜ ′′ is the centrifugal distortion
e
constant, and Y ) A1/B˜1′′. The equation employed to calculate
the rotational energy in the X2Π3/2 spin-state (F2 manifold) was
ν˜(J′′)F
)
2
1 2
2
1
1 2
2
J′′ +
- 1 +
4 J′′ +
+ Y(Y - 4) B˜′′-
[
(
]
1
)
(
)
2x
(J′′ + 1)4D˜ ′′ (3)
e
where J ) N - 1/2. The rotational distribution of the NO(X2ΠΩ,

Photofragmentation of [ClNO]n van der Waals Clusters
J. Phys. Chem. A, Vol. 103, No. 13, 1999 1931
υ′′ ) 1, J′′) states emanating from the photofragmentation of
ClNO at 355 nm was represented by a Gaussian type function
2
J′′ - J˜′′
D(J′′)
G(J′′) ) SJ′′,J′ exp -4 ln 2
(4)
(
)
[
]
with a full width at half-maximum (fwhm) ∆(J′′) ) 17 and a
mean rotational state J′′ ) 34.5. The line strength factors for
the A2Σ+ - X2ΠΩ transitions are denoted by SJ′′,J′ and we
employed the formula available from the literature for Hund’s
intermediate case between (a) and (b) because we are detecting
high NO rotational states.26 On the other hand, the distributions
of rotational states of NO emanating from the ClNO + Cl
reaction as well as NO relaxed due to its interaction with the
clusters were parametrized using a Boltzmann function. The
line shapes were represented using a Gaussian function with a
constant fwhm of 0.4 cm-1. The computer code was tested using
a laser excitation spectrum of thermal NO at 100 mTorr. No
aligment effects were included in our analysis. We believe this
should not change our qualitative interpretations.
Figure 1. Distribution of rotational states of NO(2ΠΩ, υ′′ ) 1). The
ratio of the spin-orbit state was F2/F1 ) 1.2, and the Λ-doublet state
relative populations Π(A′′)/Π(A′) were 2.0 and 4.0 for the F1 and F2
manifolds, respectively.
into the translational degrees of freedom of the Cl and NO
photofragments.35 Accordingly, the NO rotational lines are
broad.27 The analysis of the photofragmentation dynamics of
ClNO is simplified by rotationally cooling the parent molecule
(JClNO ≈ 0) because the orbital angular momentum of the
electronically excited Cl-NO collision complex, LCl-NO, is
approximately equal in magnitude (neglecting the photon and
the Cl atom angular momenta) but has opposite orientation with
III. Results and Discussion
A. Description of the Photofragmentation of ClNO at 355
nm. The conclusions that one can reach about the dynamics of
the photofragmentation of [ClNO]n clusters, involving ClNO
as the light absorbing species, strongly depend on the previous
knowledge of the dynamics of the photofragmentation of bare
ClNO at 355 nm. In accordance, we summarize here the most
relevant findings about ClNO photofragmentation.
respect to the rotational angular momentum of the NO fragment
2
J
NO (i.e., LCl-NO ≈ -JNO).37 The expression P |µ‚ꢀ| dictates
that the NO photofragment is most likely ejected along the
direction of the laser polarization vector ꢀ in the laboratory frame
because VNO is parallel to µ. If one takes the direction of
propagation of the probe laser kpr as the quantization axis of J
(the space-fixed Z axis) the NO molecules traveling along kpr
in a J angular momentum state can only exist in the Mpr ) 0
substate, although the 2J + 1 substates are populated for any
other recoil directions.38
Nitrosyl chloride presents an absorption spectrum in the 200-
600 nm range consisting of several dissociative singlet and triplet
states.27,28 The photodissociation cross-section of ClNO (Do )
13 030 cm-1) at 355 nm (E ) 28 169 cm-1) is, σ ) 1.36 ×
10-19 cm2, and the quantum yield for the formation of NO and
Cl equals unity.29,30 Bai et al. found that in the photodissociation
of ClNO, ClNO + hν f NO(2ΠΩ, υ′′, J′′) + Cl(2Pj), the spin-
orbit states in which Ω equals j are preferentially produced, as
confirmed by Chichinin detecting the individual j states of
chlorine and by Huber and co-workers detecting the NO
molecule.27,31-33 In particular, the photolysis of ClNO at 355
nm has the propensity to produce the Cl(2P1/2) spin-orbit state,
which is above Cl(2P3/2) by 881 cm-1, and the NO(2Π1/2) spin-
B. Photofragmentation of ClNO at 355 nm under Bulk
Conditions. We carried out a study of the photofragmentation
of ClNO at 355 nm under bulk conditions to understand the
role played by the photolysis of the ClNO monomer in our
experiments. We centered our analysis mainly on υ′′ ) 1 instead
of υ′′ ) 0 to minimize the detection of thermal NO, yet there
is more overlap among the rotational branches in υ′′ ) 1. In
fact, a detailed study of the υ′′ ) 0 level at the same wavelength
has been reported by Reisler and co-workers.27 We derived the
relative populations of the Q1, R1, Q2, and R2 branches from
the laser excitation spectrum of NO(υ′′ ) 1) produced from
the excitation of 150 mTorr of ClNO at 355 nm, and the result
is displayed in Figure 1. The spin-orbit state population ratio
was F1/F2 ) 1.2, while the ratio of the Λ-doublet state
populations Π(A′′)/Π(A′) were 2.0 and 4.0 for the F1 and F2
manifolds, respectively. The distribution of rotational states was
parametrized using a Gaussian function, as defined in the
Experimental Section, with the spin-orbit and Λ-doublet state
preferences specified above. The Gaussian component extends
from N equals 16 to 60, peaking at N ) 34, with a fwhm equal
to 17. We observed a small contribution of thermal NO in υ′′
) 1 below N ) 16, centered around N ) 8, although it does
not appreciably overlap with the Gaussian component and does
not affect our analysis of the photofragmentation process. A
300 K Boltzmann distribution was used to account for the small
amount of thermal NO present, which we confirmed by scanning
the dye laser while blocking the photolysis beam. Bai et al.
determined a Gaussian rotational distribution with a mean N
1
orbit states (i.e., j ) Ω ) /2). The NO(2Π3/2) manifold is 123
2
cm-1 above the Π1/2 manifold. A recent examination of the
UV photochemistry of ClNO monitoring the fine-structure
branching ratio of the Cl(2 Pj) atoms suggests the breakdown
of this rule, due to nonadiabatic effects, when probing some
regions of the ClNO potential energy surfaces.34 Reisler, Huber,
and co-workers observed NO Gaussian rotational distributions
in their ClNO photodissociation studies at 355 nm.27,32 At 355
nm the energy to be partitioned between the total center-of-
mass translational energy of the photofragments and their
internal energy is 15 139 cm-1 35
On the other hand, Reisler
.
and co-workers assigned the ClNO absorption around 620 nm
to a singlet-triplet transition. They observed that the ClNO
parent molecule NO stretch evolves into NO vibrational
excitation.36
The excitation of jet-cooled ClNO with linearly polarized laser
light at 355 nm produces a spatial distribution of NO and Cl
fragments described by an anisotropy parameter â of 1.8,
consistent with a direct photofragmentation mechanism.27 Busch
and Wilson first determined that the electric dipole transition
moment µ of ClNO is oriented parallel to the Cl-N bond in
this band system and that 70% of the available energy is released

1932 J. Phys. Chem. A, Vol. 103, No. 13, 1999
Conde et al.
value of 46 in their analysis of NO(υ′′ ) 0) from the photolysis
of ClNO at the same wavelength.27
In the 2Σ+ - 2Π band system, the main-branch Q lines probe
levels of Π(A′′) symmetry, while main-branch P and R lines
probe Π(A′) states.39,40 As Figure 1 shows, the photodissociation
process produces preferentially the Π(A′′) (antisymmetric)
Λ-doublet states in both manifold and this is because the
fragmentation process is confined to a plane containing the three
atoms.39 The larger Λ-doublet state preference in the F2 manifold
explains why this manifold appears more populated. This
experiment also establishes that there is no significant relaxation
of the NO photofragment working at a ClNO pressure of 150
mTorr detecting NO with a delay time of 50 ns.
The Λ-doublet and spin-state preferences observed from the
photodissociation of ClNO are fingerprints that will allow us
recognize the presence of the NO photofragment under expan-
sion conditions at which clusters are formed and to identify
relaxation processes.
C. Dynamics of the Cl + ClNO Reaction under Bulk
Conditions. Chlorine atoms are produced during the laser
irradiation of the [ClNO]n clusters, and an aim of the present
work is to determine whether the Cl + ClNO reaction takes
place within these clusters. To this end, we studied this reaction
under bulk conditions, since little is known about its dynam-
ics.41-45
Figure 2. NO(2Σ+, υ′)0, J′ r 2ΠΩ, υ′′ ) 1, J′′) from the Cl + ClNO
reaction: (A) experimental spectrum; (B) simulated fine-structure
spectrum. A 2:1 mixture of Cl2/ClNO was photolyzed at 355 nm at a
total pressure of 150 mTorr. The rotational distribution was parametrized
using a Boltzmann function with a “temperature” of 950 K.
Grimley and Houston observed the participation of the Cl +
ClNO reaction in their photolysis studies of ClNO in the 480-
520 nm wavelength region, detecting infrared (IR) chemilumi-
nescence from NO(υ′′ ) 1).43 These workers found evidence
that supports the following mechanism,
ClNO + hν (visible light) f Cl + NO
Cl + ClNO f Cl2 + NO
(5)
(6)
Figure 3. Rotational distributions for the main-branch R and Q
transitions of NO(2ΠΩ, υ′′ ) 1), produced by the Cl + ClNO reaction,
derived from the simulation presented in Figure 2B.
When the photolysis was carried out with visible light, they
noted that reaction 5 did not contribute to the IR emission, as
NO is mainly formed in υ′′ ) 0, and measured the rate constant
for the secondary process to be (5.40 ( 0.47) × 10-12 cm3
molecules-1 s-1, monitoring NO(υ′′ ) 1). In contrast, they
observed NO(υ′′ ) 1) instantaneously from reaction 5 when
ClNO was irradiated at 355 nm. The Cl + ClNO reaction was
used by Clyne et al. to measure the Cl atom concentration in
discharge flow systems.44
account for the small amount of ClNO photolyzed. The spin-
state and Λ-doublet state preferences incorporated to the NO-
(υ′′ ) 1) Gaussian component were the same ones previously
specified for the NO photofragment. The populations of the main
branch R and Q transitions derived from the simulation are
plotted in Figure 3.
We investigated the Cl + ClNO reaction under bulk condi-
tions using a 2:1 mixture of Cl2 and ClNO to favor the photolysis
of Cl2 as the Cl atom source. Although the magnitude of the
absorption cross sections at 355 nm of Cl2 and ClNO are similar,
we used an excess of Cl2 to favor reaction 6, which combines
with the fact that the Cl2 photodissociation produces two
Cl(2P3/2) atoms. On the other hand, the ClNO photodissociation
at 355 nm contributes Cl(2P1/2) atoms, but with a much broader
distribution of translational energies.
A laser excitation spectrum of NO(υ′′ ) 1) from the
photolysis of the Cl2/ClNO mixture at a total pressure of 150
mTorr is presented in Figure 2A. A Boltzmann rotational
distribution with a “temperature” of 950 ( 100 K describes
this ensemble of NO molecules, which was determined from
the simulation presented in Figure 2B. The “rotational temper-
ature” allows the calculation of the average energy disposed
by the reaction into NO(υ′′ ) 1) rotational energy, Erot(NO) ≈
660 cm-1. The reaction also presents a small Λ-doublet state
preference of Π(A′′)/Π(A′) ) 1.2. The relative population of
the spin-orbit states is F1/F2 ) 2.5, as can be noted from Figure
3. A Gaussian component was also used in the simulation to
D. Observation of NO(υ′′ ) 1) from the Photofragmen-
tation of [ClNO]n Clusters. The NO rotational distributions
observed during the laser irradiation at 355 nm of a 5% mixture
of ClNO diluted in He, expanded at backing pressures below 1
bar, are similar to the one presented in Figure 1, although the
maximum is slightly shifted toward larger N values. As an
example, the NO(υ′′ ) 1) rotational distributions obtained by
working at a backing pressure of 1 bar are shown in Figure 4.
We find the same Λ-doublet state and spin-orbit state prefer-
ences as in our photolysis studies of ClNO under bulk conditions
(see Figure 1). This shows that still at 1 bar there is no
significant ClNO clustering.
When the backing pressure of the mixture is increased to 3
bar, there appears a Boltzmann ensemble of NO(υ′′ ) 1)
molecules described by a “temperature” of 310 ( 30 K and no
spin-orbit preference, overlapped with the Gaussian distribution
already assigned to the NO photofragment. This is the lowest
backing pressure at which the Boltzmann component is ob-
served. We derived the values of the rotational temperature and
the spin-orbit populations from a spectral simulation, and the
distributions are displayed in Figure 5. The Gaussian component

Photofragmentation of [ClNO]n van der Waals Clusters
J. Phys. Chem. A, Vol. 103, No. 13, 1999 1933
Figure 4. Rotational distributions for the main-branch R and Q
transitions of NO(2ΠΩ, υ′′ ) 1) from the photofragmentation of jet-
cooled ClNO monomers at 355 nm. A 3% mixture of ClNO diluted in
He was expanded at a stagnation pressure of 1 bar.
Figure 6. Boltzmann plots constructed from the integrated areas of
rotational lines of NO in υ′′ ) 0. A ClNO/He sample was expanded at
a backing pressure of 3 bar and irradiated at 355 nm. (A) The “rotational
temperature” was 174 ( 26 K for the F1 spin-orbit state. (B) The
“rotational temperature” was 163 ( 25 K for the F2 spin-orbit state.
backing pressures ranging from 3 to 6 bar, showing a gradual
relative reduction of the Gaussian component with respect to
the Boltzmann one. An increase in the backing pressure
translates into an increase in the average cluster size, which
increases the probability of relaxation of the NO photofragment.
Likewise, it should be more likely to observed the reaction upon
increasing the backing pressures.
Figure 5. Rotational distributions for the main-branch R and Q
transitions of NO(2ΠΩ, υ′′ ) 1), from the photolysis at 355 nm of a
3% mixture of ClNO diluted in He and expanded at a backing pressure
of 3 bar.
E. Observation of NO(υ′′ ) 0) from the Photofragmen-
tation of [ClNO]n Clusters. The distribution of NO(υ′′ ) 0)
rotational states from the irradiation of a sample expanded at a
backing pressure of 3 bar was determined (1) from the spectral
simulation of the NO experimental spectrum and (2) by
integrating individual rotational lines. Only lines from different
main branches with no overlap were considered and the
integrated areas were corrected for line strength factors, although
not for alignment effects. Following this approach we deter-
mined the “rotational temperatures” for the F1 and F2 manifolds
to be 174 ( 26 and 163 ( 25 K, respectively (Figure 6). Using
an average temperature of 170 K, we obtained excellent
agreement between the simulated spectrum and the experimental
spectrum. The rotational distributions of NO(υ′′ ) 0) obtained
following these procedures are colder than that for NO(υ′′ )
1) detected at the same backing pressure, and this implies that
the former ensemble of molecules has been subjected to stronger
relaxation conditions. The Boltzmann ensemble of NO mol-
ecules in υ′′ ) 0 may originate from either or both of two
sources: the fraction of ClNO parent molecules photolyzed in
the interior of the clusters, where they can relax more effectively,
and/or the Cl + ClNO reaction. While the reaction studied under
bulk conditions exhibits strong spin-orbit state preference and
a rotational temperature of 950 ( 100 K, the rotational
temperature observed under cluster conditions is about 800 K
lower, and the spin-orbit states are statistically populated. That
is to say, if the reaction is contributing to the Boltzmann
has the same parameters as those employed in the simulation
used to obtain Figure 4, which implies that the NO photofrag-
ment is being detected without undergoing collisional relaxation.
This ensemble of NO molecules may originate from the
photolysis of bare ClNO and/or NO molecules ejected from the
surface of the clusters. Because the Boltzmann distribution was
310 K (i.e., close to room temperature), we carefully reduced
as much as possible the length of the gas pulses, corroborated
that the signal was observed only when the photolysis beam
was entering the chamber, and adjusted a delay time of 50 ns
between the photolysis and probe lasers. The lower NO
rotational temperature observed under cluster conditions, as
compared to 950 K observed from the Cl + ClNO reaction, as
well as the absence of spin-orbit state preference, indicate that
NO has been relaxed if the reaction is the source of this new
ensemble of molecules.
The relative contribution of the NO(υ′′ ) 1) photofragment
to the spectrum is drastically reduced upon increasing the
backing pressure to 6 bar. However, the relaxation conditions
in the jet environment are not strong enough to vibrationally
relax NO(υ′′ ) 1). The spectral region close to the band heads
was parametrized using a Boltzmann function with a “temper-
ature” of 325 ( 35 K. Neither spin-orbit state nor Λ-doublet
state preferences are observed under these conditions. Nitric
oxide is not part of the [ClNO]n complexes during its detection
because otherwise we should not be observing the NO(υ′′ ) 1)
fine structure. Other spectra (not shown) were collected at

1934 J. Phys. Chem. A, Vol. 103, No. 13, 1999
Conde et al.
component, then the NO product is being subjected to strong
relaxation conditions.
Huber and co-workers observed NO bimodal distributions
in their study of the photofragmentation of [CH3ONO]n clusters
as well as Ning and Pfab from the photofragmentation of
HONO/water heterogeneous clusters.20,22 Both groups assign the
Boltzmann component to nitrous oxide photofragments that have
been relaxed. In their experiments, the nascent distributions of
the NO photofragment are Gaussian, as in our experiments,
although there are no reactions that can produce NO. Thus, their
findings stress the importance of considering relaxation pro-
cesses within the cluster environments in our experiments.
F. Kinematic Analysis of the ClNO Photofragmentation.
After the absorption of a photon by the ClNO molecule, the
available energy to the Cl and NO products is Eavl ) Eint(ClNO)
+ hν - Do(ClNO) ) Eso(Cl) + Ekin(Cl) + Eint(NO) + Ekin(NO),
where Eint(ClNO) is the internal energy of ClNO (which may
be neglected in studies of jet-cooled ClNO or approximated as
the rotational energy of ClNO in gas-phase experiments at room
temperature), hν is the photon energy, Do(ClNO) is the
dissociation energy of ClNO, Ekin(Cl) and Ekin(NO) are the
corresponding kinetic energies of the Cl and NO fragments,
Eso(Cl) is the spin-orbit energy of the Cl(2P1/2) atom (prefer-
entially formed during the photofragmentation of ClNO at 355
nm), and Eint(NO) is the internal energy of NO. Conservation
of linear momentum in the center-of-mass relates Ekin(Cl) to
E
kin(NO), while spectroscopically one can scan individual vibro-
Figure 7. (A) NO spectrum from the photolysis of a 1:1 mixture of
Cl2:ClNO photolyzed under bulk conditions at a pressure of 150 mTorr.
The Cl + ClNO reaction and the ClNO photodissociation are observed
simultaneously. Figures B-E display NO spectra recorded under jet
conditions, at the following backing pressures: (B) 1.7 bar, (C) 2.3
bar, (D) 3.0 bar, and (E) 4.4 bar. Spectrum B corresponds to the
photodissociation of ClNO with no significant clustering. The relative
rotational states of NO and assign Eint(NO). It follows that for
a given set of quantum numbers, the NO velocity is uniquely
determined,
2(Eav1 - Eint(NO))
υ
NO(Fi, υ′′, J′′) )
(7)
S
intensity of the R1(23.5) + R21(18.5) line increases upon increasing
mNO(1 + mNO/mCl)
x
the backing pressure because the Cl + ClNO reaction and relaxation
processes are simultaneously taking place.
where mNO and mCl are the atomic masses of the specified
species. Once the velocity of the NO photofragment has been
calculated, the Doppler width is readily estimated using eq 8
for a rotational line,
and from the ClNO photodissociation. Qualitatively, the P1(36.5)
S
line is wider than the R1(23.5) + R21(18.5) line because the
former probes the ClNO photofragmentation, whereas the latter
probes the NO reaction product. On the other hand, parts B-E
of Figure 7 display NO spectra recorded under jet conditions,
increasing the backing pressure from 1.7 to 4.4 bar. Figure 7B
corresponds to the photodissociation of ClNO with no significant
clustering. Since the P1(36.5) line probes the fragmentation
process almost exclusively, the spectra depicted in Figure 7C-E
have been normalized with respect to it. It is noted that the
υ
NO(Fi, υ′′, J′′)
∆ν˜Doppler ) 2ν˜o
(8)
c
where ∆ν˜Doppler is the fwhm of the rotational line, c is the speed
of light, and ν˜o is the center frequency. For Gaussian-shaped
lines, the Doppler width is obtained by knowing the laser
bandwidth, ∆ν˜Doppler ) ∆ν˜Exp - ∆ν˜Laser2, where ∆ν˜Exp is the
experimental fwhm of the lines. Because ∆ν˜Laser is 0.04 cm-1
for our dye laser system, the correction due to ∆ν˜Laser to the
Doppler widths is very small. Employing eq 8, we can estimate
a fwhm of 0.39 cm-1 for the NO(2Π3/2, υ′′ ) 1, J′′ ) 38.5)
vibro-rotational state, where the center frequency is 42 624
intensity of the R1(23.5) + R21(18.5) line increases upon
2
2
S
increasing the backing pressure because the Cl + ClNO reaction
and relaxation processes are taking place. It is also noted that
P
the P2(38.5) + Q12(38.5) line is subjected to a relaxation
process, although it probes the photofragmentation process.
Figure 8A displays the Doppler profile of the R1(23.5) +
SR21(18.5) line recorded under the conditions specified in Figure
7A; namely, both the reaction and the photofragmentation
process are being observed simultaneously in the gas phase.
This line has been fitted by adding two Gaussian functions
centered at 42 623 cm-1: c1G1 + c2G2, with c1 ) 0.70 and
∆ν˜1 ) 0.17 cm-1, and c2 ) 0.30 and ∆ν˜2 ) 0.38 cm-1. That is
to say, this lines appears in a region of the spectrum in which
both rotational distributions overlap. As a result of the particular
conditions of this experiment, there is 40% more contribution
from the reaction to this line.
cm-1
.
G. Translational Energy Measurements. Both the photoly-
sis of ClNO at 355 nm and the Cl + ClNO reaction produce
NO(υ′′ ) 0, 1) and the information provided so far is not
sufficient to conclude whether the Cl + ClNO reaction is taking
place within the [ClNO]n clusters. Fortunately, Doppler spec-
troscopy enabled us to measure the translational energy of NO,
allowing us to identify the different processes occurring within
the cluster.
Figure 7A displays part of the NO spectrum from the
photolysis of a 1:1 mixture of Cl2:ClNO photolyzed under bulk
conditions at a total pressure of 150 mTorr. In this experiment
NO is observed simultaneously from the Cl + ClNO reaction
The P2(38.5) + PQ12(38.5) line is depicted in Figure 8B, with
fwhm of 0.38 cm-1, recorded during the photolysis of 150 mTorr
of ClNO. This line probes the NO(2Π3/2, υ′′ ) 1, J′′ ) 38.5)

Photofragmentation of [ClNO]n van der Waals Clusters
J. Phys. Chem. A, Vol. 103, No. 13, 1999 1935
(23.5) + SR12(18.5) line we obtain Ttrans ≈ 1855 K, which allows
us to calculate the mean velocity and the mean kinetic energy
to be 1144 m/s and 1640 cm-1, respectively.
The width and Doppler profile of the R2(8.5) line (Figure
8D) are the same regardless of whether this line probes the
photolysis of a sample of ClNO expanded at a backing pressure
of 3 bar or the Cl + ClNO reaction under bulk conditions. This
observation then supports the possibility that the reaction is
occurring within the clusters. For the NO reaction product
probed by the R2(8.5) line (fwhm ) 0.18 and ν˜o ) 42 445 cm-1
)
a “translational temperature” of 1052 K may be calculated by
using eq 9, together with a (lab frame) mean velocity of 862
m/s.
H. Energy Disposed into Cl2 Reaction Product. The
photodissociation of Cl2 at 355 nm is our source of Cl(2P3/2
)
atoms (â ) -1).47 The excess energy, given by Eexc ) hν +
Eint(Cl2) - Do(Cl2), can be estimated to be 8170 cm-1 if the
internal energy of Cl2 is neglected (which is mostly rotational
at room temperature). It follows that the velocity of the Cl atoms
in the center-of-mass frame is wCl ) 1659 m/s.
The available energy for the Cl + ClNO reaction is given by
Eavl ) ∆H° + Ecoll + Eint(ClNO)
) Eint(NO) + Eint(Cl2) + Etrans(NO) + Etrans(Cl2)
(10)
where ∆H° ) -6982 cm-1 is the enthalpy of the reaction, Ecoll
is the collision energy, and Eint and Etrans denote the internal
energies and translational energies of the indicated species,
respectively. Although the collision energy is given by
S
Figure 8. (A) Doppler profile of the R1(23.5) + R21(18.5) line
recorded under the conditions specified in Figure 7A. This line has
been fitted by adding two Gaussian functions centered at 42 623 cm-1
:
c1G1 + c2G2, with c1 ) 0.70 and ∆ν˜1 ) 0.17 cm-1 and c2 ) 0.30 and
1
2
2
Ecoll ) m(wCl2 + υ 2 + υClNO
)
(11)
∆ν˜2 ) 0.38 cm-1. (B) The P2(38.5) + Q12(38.5) line with fwhm of
P
Cl2
0.38 cm-1, recorded during the photolysis of 150 mTorr of ClNO. The
Doppler width of this line allows one to calculate the (lab frame)
velocity of the photofragment to be 1336 m/s. (C) The R1(23.5) + SR12-
(18.5) line (fwhm ) 0.24 cm-1, and νo ) 42623 cm-1) detected at a
backing pressure of 3 bar. Its Doppler profile is Gaussian and probes
a Maxwell-Boltzmann distribution of velocities to which can be
associated a translational temperature Ttrans ≈ 1855 K, a mean velocity,
and a mean kinetic energy of 1144 m/s and 1640 cm-1, respectively.
(D) The NO reaction product probed by the R2(8.5) line (fwhm ) 0.18
and ν˜o ) 42 445 cm-1) with a “translational temperature” of 1052 K
and a (lab frame) mean velocity of 862 m/s. The width and Doppler
profile of this line are the same regardless whether this line probes the
photolysis of a sample of ClNO expanded at a backing pressure of 3
bar or the Cl + ClNO reaction under bulk conditions.
it may be approximated as (µwCl2)/2 ≈ 2780 cm-1 (33.2 kJ/
mol), where µ is the reduced mass of the Cl and ClNO reactants.
An estimate of Eint(ClNO) of 310 cm-1 is obtained by assuming
that most of the ClNO internal energy is rotational. Therefore,
the energy available to the reaction amounts to 10 072 cm-1
.
The thermal distribution of internal states of the Cl2 molecule
(the Cl atom source) introduces considerable smearing into the
translational energy distribution of the Cl atoms, which is
described by a Gaussian function with a fwhm given by48-50
4mClNO(2mCl + mClNO
)
δEcoll
)
RTEexc ln 2 (12)
2
(mCl + mClNO
)
x
vibro-rotational state for which we previously calculated a
Doppler width of 0.39 cm-1 using eqs 7 and 8. The excellent
agreement between the measured and the estimated line widths
confirms that the NO photofragment is not undergoing a
significant translational relaxation under these experimental
conditions. Moreover, the Doppler width of this line allows one
to calculate the (lab frame) velocity of the photofragment to be
1336 m/s.
For the Cl + ClNO reaction δEcoll equals 2723 cm-1 (32.6 kJ/
mol), which means that in our experiments δEcoll and Ecoll are
of the same magnitude.
We now seek to calculate the internal energy of the Cl2
(unobserved) product that correlates with a given vibro-rotational
state of NO. This can be done employing eq 10. The NO
velocity (VNO) in the lab frame is related to the center-of-mass
velocity, Vcm, and wNO by the equation VNO ) Vcm + wNO. In
the present case, Vcm ≈ mCl/(mCl + mClNO)VCl amounts to 530
m/s. Assuming that the ClNO target molecule is initially
stationary in the center-of-mass coordinate, the velocity of NO
in the lab frame is then related to Vcm and wNO by the cosine
A closer examination to the R1(23.5) + SR12(18.5) line (fwhm
) 0.24 cm-1, and ν˜o ) 42 623 cm-1) is given in Figure 8C,
which was detected at a backing pressure of 3 bar. Its Doppler
profile is Gaussian, and this line represents a Maxwell-
Boltzmann distribution of velocities to which a translational
temperature can be associated,46
2
2
2
law VNO ) Vcm + wNO + VcmwNO cos θ, where θ is center-
of-mass scattering angle of NO. Since we did not observe vector
correlations between Vcm and wNO in our high-resolution
experiments (the Doppler profiles were Gaussian), the average
mNO
2
∆ν˜c
Ttrans
)
(9)
(
)
2ν˜o 2kB ln 2
2
2
velocity of NO in the center-of-mass frame is wNO ) υNO
2
where kB is the Boltzmann constant. Using eq 9 for the R1-
- Vcm . Because in section F we found υNO ) 862 m/s, we

1936 J. Phys. Chem. A, Vol. 103, No. 13, 1999
Conde et al.
can calculate wNO and Etrans(NO) to be 679 m/s and 579 cm-1
,
respectively. The translational energy of the Cl2 product is (mNO
/
2mCl) times the kinetic energy of NO, that is, Etrans(Cl2) ) 245
cm-1. Thus, the fraction of the total available energy which is
channeled into product relative translation is ftrans ) 0.08.
In section C we calculated Erot(NO) ≈ 660 cm-1 (frot(NO) ≈
0.06). In the present experiments we are detecting NO(υ′′ )
1), with Evib(NO) ≈ 2825 cm-1. The internal energy disposed
into the Cl2 product may be estimated by eq 10 to be 5761
cm-1. In other words, the Cl2 (unobserved) fragment formed in
coincidence with the spectroscopically selected NO(υ′′ ) 1)
molecule may be formed up to υ′ ) 10, although not
electronically excited, and fint(Cl2) ≈ 0.57.
The question now is, how is this energy being distributed
between the rotational and vibrational degrees of freedoms of
Cl2? Burn and Dainton measured a very small activation energy
(≈4 kJ/mol) for this reaction and determined that the potential
energy surface is attractive.42 On the basis of Burn and Dainton’s
findings, the fact that more than 50% of the available energy
becomes Eint(Cl2), and the modest rotational excitation observed
in the NO product, one could anticipate low Cl2 rotational
excitation and argue that the Cl + ClNO reaction proceeds
through a direct atom abstraction mechanism.42,51
Figure 9. High-resolution laser excitation spectrum of low rotational
states of NO(υ′′ ) 1) from the photofragmentation of the [ClNO]n
clusters. A ClNO/He sample was expanded at a backing pressure of 3
bar. The fwhm of the rotational lines belonging to the P1 branch are
narrow. In contrast, the widths of the rotational lines that probe the F2
manifold increase with the J quantum number, which means that the
cross sections for relaxation within the F2 manifold are larger than those
for the F1 manifold.
Another equation is needed to know the answer to this
question and we should consider the angular momentum
conservation for this reaction, as it dictates how the rotational
energy is distributed between the products: LCl + JClNO + L
which is too far to be subjected to fine-structure mixing. Thus,
our observations are consistent with this model.
J. Doppler Scans of Low Rotational States of NO(υ′′ )
1) from the [ClNO]n Photofragmentation. A high-resolution
laser excitation spectrum of NO(υ′′ ) 1) from the photofrag-
mentation of the [ClNO]n clusters is presented in Figure 9. A
ClNO/He sample was expanded at a backing pressure of 3 bar.
The fwhm of the P1(1.5) line is 0.15 cm-1, and this value
remains essentially unchanged in this branch up to the P1(17.5)
line. In contrast, the widths of the rotational lines that probe
the F2 manifold increase with the J quantum number. For
example, the fwhm of the R2(8.5) line is 0.17 cm-1, whereas
) JNO + JCl + L′, where LCl denotes the orbital angular
2
momentum of the Cl atom, J are the rotational angular momenta
of the indicated species, and L and L′ are the orbital angular
momenta of the reactants and of the product, respectively. An
average impact parameter, b ≈ 10-9 m, was estimated from
the available rate constant,43 assuming that for the present
experiments it holds the condition, k ≈ υClπb2. It follows that
the orbital angular momentum in the entrance channel, L )
µVClb ≈ 100p, is the largest source of angular momentum for
P
the fwhm of the R2(10.5) and P2(20.5) + Q12(20.5) lines are
0.23 and 0.30 cm-1, respectively. The preferential broadening
of the rotational lines corresponding to high rotational states
may be explained as follows. The NO photofragments are
produced (with larger probability) in high rotational states and
with a large translational energy. However, as a result of the
interactions within the clusters, part of the rotational and
translational energy is transferred, and this happens more
efficiently to levels closer in energy. These relaxed states are
then observed as the tail of the Boltzmann ensemble of nitric
oxide molecules. Moreover, since the lower rotational lines of
NO in the F1 manifold recorded under cluster conditions are
not undergoing a systematic broadening, they do not probe the
relaxed NO photofragment and must belong to the NO reaction
product.
the reaction. Although we measured JNO experimentally, JCl
2
+ L′ are still unresolved. We indeed know that 8% of Eavl
appears as translational energy of the product. It follows that
the relative product velocity (υ′) is small and its contribution
to the exit channel orbital angular momentum should be small
as well: L′ ) µ′b′υ′, where µ′ and b′ are the reduced mass of
the Cl2 and NO product and the exit channel impact parameter,
respectively. Thus, it is still necessary to know b′ and the
geometry of the Cl-ClNO reaction complex in order to estimate
JCl
.
2
I. Spin-Orbit Effects. As we already mentioned, the relative
population of the spin-orbit states is F1/F2 ) 2.5 for the
reaction. It should be noted that the spin-orbit splitting constant
of NO is 123 cm-1, which is 1% of the total available energy.
Therefore, a statistical population of the spin-orbit states should
From Figure 4 we observe that the NO photofragment is
preferentially formed in the F2 manifold in the antisymmetric
Λ-doublet state, as probed by the Q-branch. On the other hand,
Figure 9 shows that both the symmetric and antisymmetric
Λ-doublet states of the F2 manifold are broadened, as probed
be close to unity and clearly the reaction is producing NO(2Π1/2
,
N′′) with a market preference. We are actually observing the
reaction involving the lowest energy spin-orbit states, Cl(2P3/2
)
+ ClNO f Cl2 + NO(2Π1/2). This indicates that the reaction
Q
follows an adiabatic trajectory and there are no long-range
by the R2 and Q2 + R12 branches, respectively. Therefore, a
2
interactions that can cause transitions between the Π1/2 and
propensity rule to conserve the symmetry of Λ-doublet states
is not observed. Because the lower rotational states in the F1
manifold are not broadened, this implies that the population
transfer from the F2 manifold to the F1 manifold is less efficient
than the relaxation within the F2 manifold.
2Π3/2 states in the exit channel. Actually, the time τ required to
observe transitions between two adiabatic states is p/∆V ) 14
fs, where ∆V in the present case is the spin-orbit splitting
2
2
between the two surfaces correlating with the Π1/2 and Π3/2
states.52 Since we calculated the velocity of NO in the center-
of-mass to be 679 m/s, we can estimate that during the time τ
the NO reaction product is 28 Å away from the center-of-mass,
Our results are in qualitative agreement in two aspects with
the theoretical work of Orlikowski and Alexander on the NO-
(2Π) + Ar system.53 First, these authors found that the cross

Photofragmentation of [ClNO]n van der Waals Clusters
J. Phys. Chem. A, Vol. 103, No. 13, 1999 1937
sections for inelastic collisional transitions involving rotational
levels that belong to different manifolds are smaller than those
for transitions within a manifold. This first point is consistent
with the results displayed in Figure 9, in the sense that we
observe intramultiplet relaxation in the F2 manifold. Second,
they predicted that for transitions between specified values of
J and J′ within a manifold, the cross section for ꢀ-changing
transitions are larger for the F2 manifold. In our experiment we
do not have state-to-state resolution to rigorously confirm this
rule, but on the basis of the systematic broadening observed
to different manifolds are smaller than those for transitions
within a manifold and that the cross sections for relaxation
within the F2 manifold are larger than those for F1, and these
results are in qualitative agreement with theory.53
Acknowledgment. The financial support for this work from
the U.S. Department of Energy EPSCoR Program, by Fondos
Institucionales de la Universidad de Puerto Rico, and by the
NIH-MBRS Program, is gratefully acknowledged. The excimer
laser was purchased with funds of the Puerto Rico Laser Facility,
under the auspices of the RCMI Program and NSF-EPSCoR.
C.C. is grateful for the award of a GAANN fellowship. The
authors wish to thank Ruth Molina for the design of some parts
of the molecular beam chamber, Rudy Rivera for the construc-
tion of several parts of the molecular beam machine, and
Michael Joyner for the construction of a circuit to operate the
pulsed solenoid valve.
P
Q
for the R2, P2 + Q12, and Q2 + R12 branches, it is clear that
the cross sections for relaxation within the F2 manifold are larger
than those for F1.
IV. Summary
The photofragmentation of ClNO at 355 nm produces NO in
υ′′ ) 1 with a small spin-orbit preference of F1/F2 ) 1.2, but
with Λ-doublet state preferences of Π(A′′)/Π(A′′) ) 2.0 and
4.0 for the F1 and F2 manifolds, respectively. The distribution
of rotational states was parametrized using a Gaussian function
centered at N ) 34, with a fwhm of 17. The fwhm of the
References and Notes
(1) Castleman, A. W., Jr.; Bowen, K. H., Jr. J. Phys. Chem. 1996,
100, 12911. Echt, O.; Dao, P. D. Morgan, S.; Castleman, A. W., Jr. J. Chem.
Phys. 1985, 82, 4076. Stephan, K.; Futrell, J. H.; Peterson, K. I.; Castleman,
A. W., Jr.; Mark, T. D. J. Chem. Phys. 1982, 72, 2408.
(2) Chemical Reactions in Clusters; Bernstein, E. R., Ed.; Oxford
University: New York, 1996. Gerber, R. B.; McCoy, A. B.; Garc´ıa-Vela,
A. Annu. ReV. Phys. Chem. 1994, 45, 275.
(3) Schro¨der, T.; Schinke, R.; Liu, S.; Bacic, Z.; Moskowitz, J. W. J.
Chem. Phys. 1995, 103, 9228.
(4) Finney, L. M.; Martens, C. C. J. Phys. Chem. 1993, 97, 13477.
(5) . Droz, T.; Knochenmuss, R.; Leutwyler, S. J. Chem. Phys. 1990,
93, 4520. Knochenmuss, R.; Leutwyler, S. J. J. Chem. Phys. 1989, 91, 1268.
Knochenmuss, R.; Cheshnovsky, O.; Leutwyler, S. Chem. Phys. Lett. 1988,
144, 317.
(6) Cohen, R. C.; Saykally, R. J. J. Phys. Chem. 1992, 96, 1024. Bacic,
Z.; Miller, R. E. J. Phys. Chem. 1996, 100, 12945. Willey, K. F.; LaiHing,
K.; Taylor, T. G.; Duncan, M. A. J. Phys. Chem. 1996, 97, 7435.
(7) Chem. ReV. 1994, 94 (issue of van der Waals Molecules), 1721.
Buck, U. J. Phys. Chem. 1988, 92, 1023. Nesbitt, D. J. Chem. ReV. 1988,
88, 843.
rotational lines of the NO photofragment is ≈0.38 cm-1
,
corresponding to a kinetic energy of 2240 cm-1
.
The Cl + ClNO reaction was studied under bulk conditions
at a collision energy of 2780 cm-1, and the total available energy
was 10 070 cm-1. The distribution of rotational states for the
υ′′ ) 1 level of NO is described by a Boltzmann function with
a temperature of 950 ( 100 K, corresponding to an average
rotational energy of 660 cm-1. The relative population of the
spin-orbit states was F1/F2 ) 2.5, with a small Λ-doublet state
preference of Π(A′′)/Π(A′) ) 1.2. A fwhm of 0.18 cm-1 was
measured for the NO rotational lines, corresponding to a
translational temperature of 1052 K, and a (center-of-mass)
velocity and translational energy of 679 m/s and 578 cm-1
,
respectively. The Cl + ClNO reaction proceeds through a direct
atom abstraction mechanism and 57% of the total available
energy is channeled into the Cl2 internal degrees of freedom.
Only 8% of Eavl is disposed into the product relative translational
energy.
(8) Shin, S. K.; Chen, Y.; Nickolaisen, S.; Sharpe, S. W.; Beaudet, R.
A.; Wittig, C. AdV. Photochem. 1991, 16, 249. Buelow, S.; Noble, M.;
Radhakrishnan, G.; Reisler, H.; Wittig, C. J. Phys. Chem. 1986, 90, 1015.
(9) King, D. S.; Sauder, D. G.; Casassa, M. P. J. Chem. Phys. 1992,
97, 5919. King, D. S.; Sauder, D. G.; Casassa, M. P. J. Chem. Phys. 1994,
100, 4200.
(10) Breckenridge, W. E. Acc. Chem. Res. 1989, 22, 21.
(11) Duval, M. C.; Soep, B.; Breckenridge, W. E. J. Phys. Chem. 1991,
95, 7153.
(12) Shorter, J. H.; Casassa, M. P.; King, D. S. J. Chem. Phys. 1988,
88, 4739.
(13) Honma, K.; Fujimura, Y.; Kajimoto, O.; Inoue, G. J. Chem. Phys.
1988, 88, 4739. Bai, Y. Y.; Ogai, A.; Qian, C. X. W.; Iwata, L.; Segal, G.
A.; Reisler, H. J. Phys. Chem. 1995, 99, 15573.
During the irradiation of the [ClNO]n clusters at 355 nm, a
Boltzmann ensemble of NO(υ′′ ) 1) molecules described by a
“temperature” of 310 ( 30 K with no spin-orbit or Λ-doublet
preferences is observed, overlapped with the Gaussian distribu-
tion already assigned to the NO photofragment. The Gaussian
ensemble of NO molecules may originate from the photolysis
of bare ClNO and/or NO molecules ejected from the surface of
the clusters. While the relative contribution of the NO(υ′′ ) 1)
photofragment to the spectrum is drastically reduced upon
increasing the backing pressure, the relaxation conditions in the
jet environment are not strong enough to vibrationally relax NO-
(υ′′ ) 1). Nitric oxide is not part of the [ClNO]n complexes
during its detection. The distribution of NO(υ′′ ) 0) rotational
states was Boltzmann with “rotational temperatures” for the F1
and F2 manifolds of 174 ( 26 and 163 ( 25 K, respectively.
The Boltzmann ensemble of NO molecules in υ′′ ) 0 may
originate from either or both of two sources: the fraction of
ClNO parent molecules photolyzed in the interior of the clusters,
where they can relax more effectively, and/or the Cl + ClNO
reaction. During the photofragmentation of the clusters, the NO
photofragment is translationally and rotationally relaxed. Our
high-resolution experiments provide evidence that supports that
the Cl + ClNO reaction is taking place in the environment of
the clusters. We find that the relative cross sections for inelastic
collisional transitions between rotational levels which belong
(14) Levy, D. H. AdV. Chem. Phys. 1981, 47, 323.
(15) Ewing, G. E. J. Phys. Chem. 1987, 91, 4662.
(16) Drobits, J. C.; Skene, J. M.; Lester, M. I. J. Chem. Phys. 1986, 84,
2896.
(17) Cline, J. I.; Sivakumar, N.; Evard, D. D.; Bieler, C. R.; Reid, B.
P.; Halberstadt, N.; Hiar, S. R.; Janda, K. C. J. Chem. Phys. 1989, 90,
2605. Butz, K. W.; Catlett, D. L.; Ewing, G. E.; Krajnovich, D.; Parmenter,
C. S. J. Phys. Chem. 1986, 90, 3533.
(18) Garvey, J. F.; Bernstein, R. B. J. Am. Chem. Soc. 1986, 108, 6096.
Garvey, J. F.; Bernstein, R. B. J. Phys. Chem. 1986, 90, 3577. Garvey, J.
F.; Bernstein, R. B. Chem. Phys. Lett. 1986, 126, 394.
(19) Sivakumar, N.; Burak, I.; Cheung, W.-Y.; Houston, P. L.; Hepburn,
J. W. J. Phys. Chem. 1995, 89, 3609. Sivakumar, N.; Hall, G. E.; Houston,
P. L.; Hepburn, J. W.; Burak, I. J. Chem. Phys. 1988, 88, 3692.
(20) (a) Bergmann, K.; Huber, J. R. J. Phys. Chem. A 1997, 101, 259.
(b) Kades, E.; Ro¨sslein, U.; Bru¨hlmann, U.; Huber, J. R. J. Phys. Chem.
1993, 97, 989.
(21) Jones, A. B.; Lopez-Martens, R.; Stace, A. J. J. Phys. Chem. 1995,
99, 6333.
(22) Ning, C. L.; Pfab, J. J. Phys. Chem. A 1997, 101, 6008.
(23) Rudich, Y.; Hurwitz, Y.; Lifson, S.; Naaman, R. J. Chem. Phys.
1993, 98, 2936.
(24) Langridge-Smith, P. R. R.; Carrasquillo, E.; Levy, D. H. J. Chem.
Phys. 1981, 74, 6513.

1938 J. Phys. Chem. A, Vol. 103, No. 13, 1999
Conde et al.
(25) Huber, K. P.; Herzberg, G. Molecular Spectra and Molecular
Structure, IV. Constants of Diatomic Molecules; Van Nostrand Reinhold:
New York, 1979. Herzerg, G. Spectra of Diatomic Molecules; Van Nostrand
Reinhold: New York, 1950; p 222.
(26) Schleicher, D. G.; A’Hearn, M. F. Astrophys. J. 1982, 258, 864.
(27) Bai, Y. Y.; Ogai, A.; Qian, C. X. W.; Iwata, L.; Segal, G. A.;
Reisler, H. J. Chem. Phys. 1989, 90, 3903.
(28) Vegiri, A.; Alexander, M. H. J. Chem. Phys. 1994, 101, 4722.
Schinke, R.; Nonella, M.; Suter, H. U.; Huber, J. R. J. Chem. Phys. 1990,
93, 1098.
(29) Roehl, C. M.; Orlando, J. J.; Calvert, J. G. J. Photochem. Photobiol.
A.: Chem. 1992, 69, 1.
(30) Calvert, J. G.; Pitts, J. N. Photochemistry; John Wiley and Sons:
New York, 1966; p 230.
(31) Chichinin, A. I. Chem. Phys. Lett. 1993, 209, 459.
(32) Tictin, A.; Bruno, A. E.; Bru¨hlmann, U.; Huber, J. R. Chem. Phys.
1988, 125, 403.
(33) Bruno, Y. Y.; Bruhlmann, U.; Huber, J. R. Chem. Phys. 1988, 120,
155.
(34) Skorokhodov, V.; Sato, Y.; Suto, K.; Matsumi, Y.; Kawasaki, M.
J. Phys. Chem. 1996, 100, 12321.
(35) Busch, G. E.; Wilson, K. R. J. Chem. Phys. 1972, 56, 3655.
(36) Qian, C. X. W.; Ogai, A.; Iwata, L.; Reisler, H. J. Chem. Phys.
1990, 92, 4296. Qian, C. X. W.; Ogai, A.; Brandon, J.; Bai, Y.; Reisler, H.
J. Phys. Chem. 1991, 95, 6763.
(37) Hall, G. E.; Sivakumar, N.; Chawla, D.; Houston, J. W.; Burak, I.
J. Chem. Phys. 1988, 88, 3682.
(38) Greene, C. H.; Zare, R. N. J. Chem. Phys. 1983, 78, 6741. Case,
D. A.; McClelland, G. M.; Herschbach, D. R. Mol. Phys. 1978, 35, 541.
Ashfold, M. N. R.; Lambert, I. R.; Mordaunt, D. H.; Morley, G. P.; Western,
C. M. J. Phys. Chem. 1992, 96, 2938. Dixon, R. N. J. Chem. Phys. 1986,
85, 1866. Zare, R. N. Angular Momentum; Wiley: New York, 1988; p
314. Orr-Ewing, A. J.; Zare, R. N. In The Chemical Dynamics and Kinetics
of Small Radical, Liu, K., Wagner, A., Eds.; World Scientific: Singapore,
1995; Vol. 2, p 936. Dagdigian, P. J. In Atomic and Molecular Beam
Methods; Scoles, G., Ed.; Oxford University: New York, 1988; Vol. 1, p
596.
(39) Alexander, M. H.; Andresen, P.; Bacis, R.; Bersohn, R.; Comes,
F. J.; Dagdigian, P. J.; Dixon, R. N.; Field, R. W.; Flynn, G. W.; Gericke,
K.-H.; Grant, E. R. Howard, B. J.; Huber, J. R.; King, D. S.; Kinsey, J. L.;
Kleinermanns, K.; Kuchitsu, K.; Luntz, A. C.; McCaffery, A. J.; Pouilly,
B.; Reisler, H.; Rosenwaks, S.; Rothe, E. W.; Shapiro, M.; Simons, J. P.;
Vasudev, R.; Weisenfeld, J. R.; Wittig, C.; Zare, R. N. J. Chem. Phys.
1988, 89, 1749.
(40) Andresen, P.; Ondrey, G. S.; Titze, V.; Rothe, E. W. J. Chem. Phys.
1984, 80, 2548.
(41) Basco, N.; Norrish, R. G. W. Proc. R. Soc. A 1962, 268, 291.
(42) Burns, W. G.; Dainton, F. S. Trans. Faraday Soc. 1952, 48, 52.
(43) Grimley, A. J.; Houston, P. L. J. Chem. Phys. 1980, 72, 1471.
(44) Clyne, M. A. A.; Cruse, H. W.; Watson, R. T. J. Chem. Soc.,
Faraday Trans. 2 1972, 68, 153.
(45) Moser, M. D.; Weitz, E.; Schatz, G. C. J. Chem. Phys. 1983, 78,
757.
(46) Trentelman, K. A.; Kable, S. H.; Moss, D. B.; Houston, P. L. J.
Chem. Phys. 1989, 91, 7498.
(47) Matsumi, Y.; Tonokura, K.; Kawasaki, M. J. Chem. Phys. 1992,
97, 1065.
(48) Chantry, P. J. J. Chem. Phys. 1971, 55, 2746.
(49) Bernstein, R. B. Comments At. Mol. Phys. 1973, 4, 43.
(50) van der Zande, W. J.; Zhang, R.; Zare, R. N. J. Chem. Phys. 1991,
95, 8205.
(51) Levine, R. D.; Bersntein, R. B. Molecular Reaction Dynamics and
Chemical ReactiVity; Oxford University Press: New York, 1987.
(52) Graff, M. M.; Wagner, A. F. J. Chem. Phys. 1990, 92, 2423.
(53) Orlikowski, T.; Alexander, M. H. J. Chem. Phys. 1983, 79, 6006
and references therein.