H+NO2 channels in the photodissociation of HONO at 193.3 nm

Article abstract of DOI:10.1021/jp002521e

The H+NO₂ channels in the 193.3-nm photodissociation of jet-cooled HONO have been examined by using high-n Rydberg-atom time-of-flight (HRTOF) technique. Center-of-mass (CM) translational energy distribution and energy-dependent angular distribution of the photoproducts reveal that the NO₂ fragments are produced in at least three electronic states: ground X²A₁ and excited qq²B₂ and qq²B₁ (and/or C²A₂) states. The overall average CM product translational energy is 〈E_T〉 = 0.3E_avail. NO₂ fragments are highly vibrationally excited in each of these electronic states, and in particular, a long vibrational progression of NO₂ bending mode in the qq²B₂ state has been observed. Branching ratios of the NO₂ electronic states are estimated: X²A₁:qq²B₂: qq²B₁/C²A₂≈0.13:0.21:0.66. The O-H bond photodissociation of HONO from the second electronically excited singlet state B¹A′ proceeds via multiple dissociation pathways. The H+NO₂(qq²B₂) product channel is via a direct dissociation (presumably in a near-planar fragmentation geometry) and has a large translational energy release, a specific NO₂ bending vibration population (indicating a significant change of the ONO angle during dissociation), and an anisotropic product angular distribution (suggesting a short excited qq¹A′ state lifetime with respect to dissociation). The H+NO₂(X²A₁) channel could be produced from a triplet excited state (which likely has a repulsive barrier along the O-H dissociation coordinate) following intersystem crossing or from the ground state of HONO after internal conversion. The H+NO₂(qq²B₁) channel requires nonadiabatic processes in a planar geometry, while in nonplanar geometries, it can be directly produced from the HONO(qq¹A′) state, consistent with its large branching ratio.

Full text of DOI:10.1021/jp002521e

J. Phys. Chem. A 2001, 105, 1465-1475
1465
H + NO2 Channels in the Photodissociation of HONO at 193.3 nm†
Gabriel Amaral, Kesheng Xu, and Jingsong Zhang*
Department of Chemistry, UniVersity of California, RiVerside, California 92521-0403
ReceiVed: July 14, 2000; In Final Form: September 27, 2000
The H + NO2 channels in the 193.3-nm photodissociation of jet-cooled HONO have been examined by using
high-n Rydberg-atom time-of-flight (HRTOF) technique. Center-of-mass (CM) translational energy distribution
and energy-dependent angular distribution of the photoproducts reveal that the NO2 fragments are produced
in at least three electronic states: ground X˜ 2A1 and excited A˜ 2B2 and B˜2B1 (and/or C˜2A2) states. The overall
average CM product translational energy is ET ) 0.3Eavail. NO2 fragments are highly vibrationally excited
in each of these electronic states, and in particular, a long vibrational progression of NO2 bending mode in
the A˜ 2B2 state has been observed. Branching ratios of the NO2 electronic states are estimated: X˜ 2A1:A˜ 2B2:
B˜2B1/C˜2A2 ≈ 0.13:0.21:0.66. The OsH bond photodissociation of HONO from the second electronically
excited singlet state Β1Α′ proceeds via multiple dissociation pathways. The H + NO2(A˜ 2B2) product channel
is via a direct dissociation (presumably in a near-planar fragmentation geometry) and has a large translational
energy release, a specific NO2 bending vibration population (indicating a significant change of the ONO
angle during dissociation), and an anisotropic product angular distribution (suggesting a short excited B˜1A′
state lifetime with respect to dissociation). The H + NO2(X˜ 2A1) channel could be produced from a triplet
excited state (which likely has a repulsive barrier along the O-H dissociation coordinate) following intersystem
crossing or from the ground state of HONO after internal conversion. The H + NO2(B˜2B1) channel requires
nonadiabatic processes in a planar geometry, while in nonplanar geometries, it can be directly produced from
the HONO(B˜1A′) state, consistent with its large branching ratio.
Introduction
to the 2a′′(πnb,O) f 3a′′[π*(NdO)] transition [promotion of a
lone-pair 2pπ electron on the OH oxygen atom to the π*(NdO)
Nitrous acid (HONO) is an important trace species in
tropospheric chemistry.1-5 Near-ultraviolet (near-UV) photolysis
of HONO generates a significant amount of OH radicals at
sunrise after nighttime accumulation and initiates the chain
reactions in photochemical smog formation in air polluted urban
areas.1-5 Accurate photochemical kinetic and mechanistic
information of HONO is essential for understanding its roles
in atmospheric chemistry. For example, the branching ratios of
the dissociation channels OH + NO and H + NO2 are important
for addressing the OH production at sunrise and NO to NO2
conversion under conditions of low ozone concentrations.
Near-UV and UV absorption spectra of HONO consist of (i)
a diffuse structured band in the 300-390 nm region and (ii) a
broad structureless band from 270 nm to below 180 nm, peaking
at ∼210 nm.5-8 On the basis of the electronic configuration of
the ground-state HONO, ...(8a′)2(1a′′)2(9a′)2(2a′′)2(10a′)2(3a′′)0,
the first absorption band in the near-UV region has been
attributed to the 10a′(nO) f 3a′′[π*(NdO)] transition (promo-
tion of an in-plane lone-pair p electron on the terminal oxygen
atom to the first antibonding π* orbital primarily located on
the terminal NdO) and corresponds to the electronic excitation
to the first singlet excited state (A˜ 1A′′ r X˜ 1A′).8-13 This band
orbital], and corresponds to the excitation to the second singlet
excited state (B˜1A′ r X˜ 1A′), although the third singlet excited
state (C˜1A′′, via the 9a′ f 3a′′ transition) might be involved as
well.9 The electronic energy levels of HONO and their correla-
tions (in a Cs point group) to the H + NO2 and OH + NO
product channels are shown in Figure 1.9,14 The possible
9,15-17
dissociation channels are listed as follows:
HONO(X˜ 1A′) (+hν) f OH(X2Π) + NO(X2Π)
∆H0o ) 48.0 kcal/mol (1a)
f OH(A2Σ) + NO(X2Π)
∆H0o ) 140.6 kcal/mol (1b)
f H(2S) + NO2(X˜ 2A1)
∆H0o ) 77.5 kcal/mol (2a)
f H(2S) + NO2(A˜ 2B2)
∆H0o ) 105.3 kcal/mol (2b)
f H(2S) + NO2(B˜2B1)
n
consists of vibrational progression (20 ) in the ν2 mode (pri-
∆H0o ) 117.9 kcal/mol (2c)
marily the terminal NdO stretch), indicating a significant change
in the terminal NdO bond length in the A˜ 1A′′ r X˜ 1A′
transition.5c,8-13 The second absorption band has been attributed
f H(2S) + NO2(C˜2A2)
∆H0o ) 120.0 kcal/mol (2d)
† Part of the special issue “Harold Johnston Festschrift”.
* To whom all the correspondence should be sent. Fax: 909-787-4713.
E-mail: jszhang@ucrac1.ucr.edu. Also at Air Pollution Research Center,
University of California, Riverside.
Photodissociation of the HONO A˜ 1A′′ state has been exten-
sively studied.10-13,18-24 Both channels 1a and 2a are energeti-
10.1021/jp002521e CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/29/2000

1466 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Amaral et al.
an out-of-plane motion, with a change in dihedral angle and
excitation in out-of-plane torsion ν6.26,27 In addition, the vector
correlation studies have revealed that, in this heterolitic N-O
dissociation, the OH fragment is ejected with its plane of rotation
(which contains the half-filled pπ orbital) preferentially aligned
perpendicular to the recoil direction (i.e., with a propeller-type
motion).26,27 An OH(A) + NO channel has been observed at
193.3 nm with a small quantum yield of 10-5, and the OH(A)
rotational distribution can be described by a statistical dissocia-
tion model.7
The H + NO2 channels (2a-d) have not been directly
observed before. Symmetry considerations indicate that, in
coplanar fragmentation pathways (Cs point group), the A˜ 1A′′
HONO correlates adiabatically with the second electronically
excited-state NO2(B˜2B1), the B˜1A′ HONO correlates with the
first excited-state NO2(A˜ 2B2), and the C˜1A′′ HONO correlates
to NO2(C˜2A2) (shown in Figure 1). The electronic configurations
of NO2 are as follows:14,17 ...(1a2)2(4b2)2(6a1)1(2b1)0 X˜ 2A1;
...(1a2)2(4b2)1(6a1)2 A˜ 2B2; ...(1a2)2(4b2)2(2b1)1 B˜2B1; and
...(1a2)1(4b2)2(6a1)2 C˜2A2. With the plane of HONO chosen as
the symmetry plane of the NO2 fragment in the Cs point group
Figure 1. Energy level and correlation diagram of the HONO system
and energetically possible dissociation channels (assuming a coplanar
fragmentation geometry). The situations for nonplanar geometries are
also discussed in the text. Upper limits of the adiabatic energies of the
HONO A˜ 1A′′ and B˜1A′ states are derived from absorption spectra,5,27
and the energy levels of other electronically excited states are scaled
on the basis of ref 9. Energetics of the product channels are derived
from thermodynamic data in ref 15 and NO2 electronic energies from
refs 16 and 17. The energy levels are drawn to scale.
2
and thus assigned as the yz plane for NO2, the B2 is then
2
reduced to A′ and B1 to A′′ in the Cs symmetry.14 At 193.3
nm, HONO is promoted to the singlet B˜1A′ state in the spin-
conserving photoexcitation.5,7,9 As HONO(B˜1A′) correlates
adiabatically with H + NO2(A˜ 2B2) in a planar geometry,
productions of NO2(X˜ 2A1), NO2(B˜2B1), or NO2(C˜2A2) from
HONO(B˜1A′) dissociation would therefore involve nonadiabatic
processes. For example, the ground-state NO2(X˜ 2A1) correlates
adiabatically in the planar geometry with the ground-state
HONO(X˜ 1A′) and the second excited triplet b˜3A′ state, photo-
dissociation of HONO(B˜1A′) to NO2(X˜ 2A1) requires intersystem
crossing with the b˜3A′ surface via spin-orbit couplings or
internal conversion to the ground state X˜ 1A′ via vibronic
interactions. Potential energy curves of the X˜ 1A′ ground state
and singlet and triplet excited states of HONO have been
examined by ab initio calculations (restricted in the planar
geometry), and the excited-state surfaces are shown to be not
purely repulsive or have substantial energy barriers along the
H-ONO coordinate.9 When nonplanar dissociation pathways
are considered, however, the NO2(B˜2B1) product correlates
adiabatically with the B˜1A′ HONO instead, and NO2(A˜ 2B2) with
the A˜ 1A′′ and b˜3A′ states of HONO, NO2(X˜ 2A1) with X˜ 1A′ and
a˜3A′′ HONO, and NO2(C˜2A2) with C˜1A′′ HONO. Consequently,
productions of NO2 in X˜ 2A1, A˜ 2B2, and C˜2A2 from nonplanar
HONO(B˜1A′) dissociation require nonadiabatic processes.
It is difficult to identify the H + NO2 channels by using
traditional methods to monitor the NO2 products, as these
techniques are severely hindered by the large background of
NO2 that is in equilibrium with HONO under laboratory
conditions. In this work, we utilized a high-n Rydberg-atom
time-of-flight (HRTOF) technique to directly detect and study
the H-atom product channels. HONO molecules were cooled
in a supersonic jet and photodissociated at 193.3 nm. H-atom
TOF spectra were obtained and yielded product translational
energy distribution, providing nascent internal state distribution
of the NO2 fragments. This work reports the first direct
observation and measurement of the H + NO2 channels in the
UV photodissociation of HONO.
cally possible in the near-UV absorption band. The OH(X2Π)
+ NO(X2Π) channel (1a) has been identified as the major
product channel with a nearly unit quantum yield, although
recent experimental evidence indicated that the H + NO2
channel (2a) possibly has a nonnegligible branching ratio at
wavelengths shorter than its threshold of 369 nm.6,25 Channel
1a has been shown to occur via a fast vibrational predisso-
ciation.10-13,18-23 State-resolved laser-induced fluorescence (LIF)
measurements have shown that the OH fragment is rotationally
and vibrationally cold (with its half-filled pπ orbital preferen-
tially oriented in the plane of rotation) and translationally hot
(with an anisotropy parameter â ∼ -1, indicating a transition
dipole moment perpendicular to the molecular plane and a
prompt and planar dissociation pathway).18-21 The NO fragment,
however, has significant internal energy with a inverted
vibrational and rotational state population,21,22 suggesting a
strong coupling between the terminal NdO bond and the N-O
dissociation coordinate. A dissociation lifetime of ∼100 fs (20
0
3
fs for 20 and 100 fs for 20 , increasing with excitation in the
excited-state NdO stretch) has been inferred from vector
correlation studies.23 Ab initio quantum mechanical calculations
have indicated that the Franck-Condon region of the A˜ -state
potential energy surface (PES) has a potential well along the
terminal NdO coordinate and a small barrier to dissociation
along the central O-N coordinate.10-13 Classical trajectory and
quantum dynamics calculations on the A˜ state surface have
confirmed many of the experimental results.10-13
There is little information on photodissociation dynamics of
the B˜1A′ state. Only a few experiments have been carried out
at 193.3 nm and at 212, 218, and 220 nm, and they have focused
on the OH + NO channels.7,26,27 OH product vector correlation,
Doppler profiles, and energy distribution indicate a large
anisotropy and fast dissociation on the B˜ state (in the order of
femtoseconds). With this prompt dissociation mechanism, the
measured anisotropy parameter â of ∼1.3 indicates that the B˜1A′
f X˜ 1A′ transition moment lies in the molecular plane and has
a ∼35° angle with the OH recoil velocity (the dissociating N-O
bond).26,27 A nonplanar equilibrium geometry in the B˜ state has
been proposed (due to removal of an out-of-plane pπ electron
on the O atom in the π-π* transition).26,27 Upon vertical
excitation, the OH moiety of the B˜-state HONO could undergo
Experimental Section
The HRTOF technique, pioneered by Welge and co-workers,
was utilized in this study and has been described previously.28-30
A pulsed molecular beam was produced by expanding HONO

H + NO2 Channels in the Photodissociation of HONO
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1467
in Ar gas mixture (at a total pressure of 120 kPa) from a HONO
generator into the source chamber through a pulsed nozzle
(General Valve, Series-9) operating at 20 Hz with a 300 µs pulse
width. The molecular beam was differentially pumped and
collimated by a 1-mm diameter skimmer located 3.0 cm from
the nozzle; 4.6 cm further downstream, the molecular beam was
crossed with a 193.3-nm UV photolysis laser beam that was a
portion of the output from an ArF excimer laser (Lambda-Physik
EMG 101). The ArF laser beam (with typical 5-10-mJ energy)
was focused with a quartz lens of ∼70 cm focal length. This
radiation could be polarized with a 10-plate stack of quartz slides
placed at the Brewster’s angle, resulting in ∼95% polarization.
Hydrogen atoms produced from photodissociation of HONO
were excited by 121.6-nm Lyman-R radiation. This vacuum-
UV radiation was generated by tripling the 364.7-nm radiation
[doubled output from a Nd:YAG pumped dye laser (Lambda-
Physik 3002)] in Kr and was focused into the interaction region
by a MgF2 lens. The H atoms were further excited from the
22P level to a high-n Rydberg level (n ∼ 40-90) by the doubled
output from another Nd:YAG pumped dye laser (Laser Analyti-
cal System 2051). Background ions (generated by photolysis
or probe laser beams) were removed by a small negative
potential applied below the interaction region and thus were
prevented from reaching the detector. The high-n Rydberg states
were radiatively metastable and stayed highly excited for many
tens of microseconds while they drifted with their nascent
velocities to a microsphere plate detector (El-Mul Technologies).
Upon arrival at the detector, the excited atoms were efficiently
field-ionized as they passed a wire mesh, and they were then
detected as ions. The ion signals were amplified by a fast
preamplifier (Phillips Scientific 6950). TOF spectra were
recorded and averaged by using a multichannel scaler (EG&G,
Turbo MCS) and were stored on a computer. The flight distance
was calibrated by photodissociating HCl, and the nominal flight
length was 37.1 cm. The dwell time used on the MCS was 20
ns and typically 10 000 channels were recorded. The ac-
cumulated TOF spectra shown in this work usually represent
200-400 thousand laser firings.
The HONO gas sample was prepared by reaction of aqueous
solutions of NaNO2 and H2SO4 in a HONO gas generator, which
is based on the designs of Ning and Pfab31 and Marshall and
Midgeley.32 A schematic representation of this HONO generator
is shown in Figure 2. The generator was essentially a glass vessel
of ∼25-cm length and ∼ 4-cm diameter. A small compartment
(∼35 mL) was attached underneath the bottom of the reaction
vessel in case of a liquid leak through the ceramic frit due to
gas-sample pressure buildup. The reactants’ aqueous solutions
(5 M H2SO4 and 5 M NaNO2) were continuously pumped into
the generator through two separate glass inlets by a two-channel
peristaltic pump with ∼5 mL/min flow rate. A second single-
channel peristaltic pump was used to continuously remove the
waste (with a pumping speed of ∼10 mL/min) to maintain a
constant level of solution inside the generator. An extra waste
outlet was connected to a vacuum line through a two-way valve,
which was open only when pressure increased inside the vessel.
Argon carrier gas entered the HONO generator from the bottom
through the frit and bubbled through the liquid sample. The
whole system was kept at room temperature. The HONO/NO/
NO2/H2O/Ar mixture exited the HONO producer through a
liquid-droplet trap compartment, and then passed through a 5-cm
long glass drying tube filled with CaCl2 as desiccant. Some
amount of H2O was still present in the sample. Since HONO
tends to decompose catalytically on contact with metal surfaces,
the HONO gas mixture was transported by glass and Telfon
Figure 2. Schematic diagram of the HONO generator, which was based
on the designs of Ning and Pfab (ref 31) and Marshall and Midgley
(ref 32) and has incorporated our modifications.
Figure 3. Schematic diagram of the flow system of the HONO sample
into the molecular beam.
tubing and fittings and delivered right behind the pulsed nozzle
in a flow delivery system (Figure 3). This flow system (with a
constant flow rate of ∼0.1 L/min) significantly reduced the
residence time of the gas sample inside the metal nozzle body
and thus minimized the decomposition of HONO. No observable
loss of HONO was detected by near-UV absorption spectra
before and after feeding into the molecular beam nozzle source.
The composition of the gas mixture in the molecular beam was
examined by using a quadrupole mass spectrometer (UTI-100C)
downstream. HONO (m/e ) 47) was clearly identified, along
with significant amount of NO, NO2, and H2O. Production of
HONO was also confirmed by its characteristic near-UV
absorption spectrum.5 During the experiments, the concentra-
tions of HONO (0.5-1%) and NO2 in the gas mixture were
continuously monitored in a 10-cm fused silica cell at the end
of the gas flow system (after the beam nozzle source) by using
the UV spectrophotometer (Varian DM 100S). Optimum condi-
tions were achieved with a total gas pressure of 120 kPa and a
gas flow of ∼0.1 L/min.
Population ratio of trans-HONO and cis-HONO isomers (with
trans-HONO being more stable by ∼0.65 kcal/mol) is ∼3.25
at 277 K,5c and this ratio is essentially unchanged even after
supersonic expansion.23 Therefore, trans-HONO was treated as
the major HONO isomer in this study. Although HNO3 was
considered as a possible impurity from this HONO generator,31,32
it was not favored under the continuous flow conditions in our
setup and was not identified in the quadrupole mass spectrom-
eter. Additional experiment of 193.3-nm photolysis of HNO3
was carried out. The H-atom TOF spectrum of HNO3, primarily

1468 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Amaral et al.
of detection), and P2(cosθ) is the second Legendre polynomial.
Due to the small H-atom mass, center-of-mass (CM) velocities
of the H atoms (those with CM product translational energy ET
> 2 kcal/mol) are much larger than the laboratory velocity of
the parent molecule (velocity of the CM). Therefore, CM and
laboratory frames are essentially the same in this study; i.e.,
the laboratory velocity and CM velocity of the H atoms are the
same, and the angle θ, defined in the CM frame, is effectively
the same as that in the laboratory frame. Similar to eq 3, CM
differential cross-sections of the photofragments, i.e., angular-
dependent product CM translational energy distributions, are
expressed as35,36
d3σ
dΕΤ dΩ 4π
1
P(ET,θ) )
)
P(ET)[1 + â(ET) P2(cos θ)] (4)
where P(ET) is the angle-integrated product CM translational
energy distribution, i.e., the integrated differential cross-sections
over scattering angles. The P(ET,θ) distributions can be obtained
by appropriate transformation of the TOF spectra I(t,θ) (with a
factor of t3).29 Note that the â parameter can be energy-
dependent.
At magic angle (θm ) 54.7°), the TOF spectrum is indepen-
dent of â; i.e., Im(t) ) (1/4π)Ι(t). The magic-angle TOF spectrum
can be obtained from the TOF spectra at θ ) 0° (|) and θ )
90° ( ) by Im(t) ) [I|(t) + I (t)]/3. Im(t) can also be derived
from the TOF spectrum of unpolarized 193.3-nm radiation,
Figure 4. H-atom TOF spectrum of 193.3-nm photodissociation of
HONO, with the polarization E vector of the photolysis radiation: (a)
parallel to the TOF axis, |, θ ) 0°; (b) at magic angle, θ ) 54.7°; and
(c) perpendicular to the TOF axis, , θ ) 90°. The ArF laser power
used was typically 5 mJ/pulse for polarized and <10 mJ/pulse for
unpolarized. The signals have been normalized to the same laser power
(5 mJ) and laser shots. The arrow indicates the fastest possible H atom
from one-photon dissociation of HONO at 193.3 nm.
Iunpol(t), by using Iunpol(t) ) (1/4π)I(t)[1 + â(t)/4]. â(t) is
calculated from â(t) ) 2[I|(t) - I (t)]/[I|(t) + 2I (t)]. The magic-
angle TOF spectra obtained by combining the | and TOF
spectra and by correcting the unpolarized spectrum are es-
sentially identical (after normalization of the laser power), which
indicates proper treatments of the TOF spectra of different
polarized photolysis radiation, the correction for the imperfect
polarization of the polarizer, and the anisotropy parameter. The
magic-angle TOF spectrum from the average of these two
approaches is shown in Figure 4b.
due to multiphoton processes,33 had a very different shape
compared to that from HONO photodissociation (at the same
photolysis energy). Photolysis of H2O at 193.3 nm was also
studied separately. Due to the small absorption cross-section
of H2O at 193.3 nm, its photolysis would produce only a small
single peak [corresponding to H + OH(X2Π, V ) 0)] with
product translational energy ET ) 30 kcal/mol,34 and this peak
was not identified in the TOF spectrum of HONO.
The P(ET,θ)’s for the three polarization angles of the
photolysis radiation, P|(ET), Pm(ET), and P (ET), are converted
from the TOF spectra in Figure 4a-c and are shown in Figure
5a-c, respectively.29 The magic-angle Pm(ET) is proportional
to P(ET) [Pm(ET) ) (1/4π)P(ET)], and is independent of â and
θ. Thus, Pm(ET) (in Figure 5b) is treated as P(ET) for calculation
of translational energy release and relative branching ratios of
the different product channels. CM product translational energy,
ET, is related to the internal energy of the NO2 fragment,
Eint(NO2), by conservation of energy:28-30
Results
H-atom TOF spectra of 193.3-nm photodissociation of HONO
are shown in Figure 4. To correct for background, spectra were
taken with molecular beam alternately on and off while all the
other experimental conditions were kept the same. Despite a
large absorption cross-section (σ ∼ 10-18 cm2) of HONO at
193.3 nm,5a,7 the signal level of the H-atom TOF spectra was
low and required an extensive signal averaging, indicating small
branching ratios of the H + NO2 channels. TOF measurements
were carried out with the 193.3-nm photolysis radiation unpo-
larized and linearly polarized [parallel (|, θ ) 0°) and
perpendicular ( ,θ ) 90°) with respect to the TOF axis]. The
TOF spectra for | and polarization (Figure 4a,c) have been
corrected for the imperfect polarization of the stacked-plate
polarizer. Since the linearly polarized photolysis radiation
preferentially excites HONO molecules with their electronic
transition dipole moments parallel to its electric vector E,
H-atom product angular distributions can be obtained in the
polarization studies.35 The angular-dependent TOF spectra of
the photofragment are described by35
ET ) hν - D0(H-ONO) - Eint(NO2)
(5)
where hν is the photon energy and D0(H-ONO) is the O-H
bond energy. Internal energy of the HONO parent molecule is
negligible due to supersonic cooling. The P(ET) distribution thus
reflects the nascent internal energy distribution of the NO2
photofragment. The energy threshold of channel 2a is 77.5 kcal/
mol,15 corresponding to a maximum available energy ET ) 70.4
kcal/mol, as indicated in the P(ET)’s in Figure 5. The corre-
sponding ET onsets of the electronically excited NO2 channels
2b-d are also marked in Figure 5. The energy-dependent
anisotropy parameter â is calculated by using
1
4π
I(t,θ) ) I(t)[1 + â(t) P2(cos θ)]
(3)
â(ET) ) 2[P|(ET) - P (ET)]/[P|(ET) + 2P (ET)] (6)
where â is the anisotropy parameter (-1 e â e 2), θ is the
angle between the E vector of the polarized laser radiation and
the recoiling velocity vector of the H-atom product (the direction
and is shown in Figure 5d.

H + NO2 Channels in the Photodissociation of HONO
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1469
Figure 6. TOF spectra of HONO photodissociation at 193.3 nm at
different laser power. The spectra are scaled to the same maximum.
P(ET).36 Assuming constant â for the individual channels and
using â2hν ) -0.9 and â1hν ≈ 0 (the lower limit), one can solve
for P2hν(ET) and P1hν(ET). The deconvoluted components are
plotted in Figure 5b. Note that this result is based on the lower
limit â1hν ≈ 0, which gives a lower limit of the contribution of
the two-photon component. The above procedure, however,
should capture the general feature of the two-photon component,
and in particular, contribution of the two-photon photodisso-
ciation at ET > 63 kcal/mol is unambiguous, as it is the only
channel in this high-energy region.
Figure 5. Product CM translational energy distributions, P(ET)’s,
derived from the TOF spectra in Figure 4, with 193.3-nm photolysis
radiation polarization: (a) parallel to the TOF axis; (b) at magic angle;
(c) perpendicular to the TOF axis; and (d) energy-dependence anisotropy
parameter â. The P(ET) distributions have been normalized to the same
scale. Onsets of the H + NO2 channels are marked. Note that both
P(ET)’s and â have contribution from two-photon dissociation. In b,
the deconvoluted one-photon component, P1hV(ET), and two-photon
component, P2hV(ET), are plotted in dot and dash lines, respectively.
See the text for details.
The most likely source of the fast H-atoms is sequential
secondary photodissociation of OH radicals generated from
193.3-nm photolysis of HONO. OH(X2Π) + NO(X2Π) (1a) is
the major channel in the primary photodissociation, and
The H-atom TOF spectra are complicated by a small amount
of signals (starting in the region of 12-15 µs) that are faster
than the maximum kinetic energy release of the H + NO2
products (Figure 4). This fastest portion has a slope that differs
from the rising edge after 15 µs in the TOF spectra, and more
importantly, it has a distinctly different angular distribution (â
≈ -0.9, from ET ∼ 63 kcal/mol to the maximum available
energy in one-photon dissociation ET ) 70.4 kcal/mol, and to
ET > 90 kcal/mol, in Figure 5d). Single-photon dissociation of
internally hot HONO molecules is not the source of the fast H
atoms, as the observed extra energy in the fast H atoms is much
larger than the possible initial internal energy of the HONO
molecules in the supersonic beam. These fast signals are due
to multiphoton dissociation of HONO or its clusters, which is
also confirmed by photolysis power dependence (Figure 6). The
H-atom TOF spectra shown in Figure 4 were taken with the
photolysis laser power lowered to ∼5 mJ in order to reduce the
two-photon signals. In addition, the early portion of the pulsed
molecular beam was used to minimize the contribution of
clusters.
The â parameter remains at a nearly constant value of ∼ -0.9
(the value of the two-photon component) from beyond 90 kcal/
mol to ∼63 kcal/mol; it then gradually increases from ∼ -0.9
at ET ∼ 63 kcal/mol to ∼0 at ∼43 kcal/mol (Figure 5d). This
change in â suggests that there are two product channels in
this region. Since H + NO2(X˜ 2A1) (channel 2a) is the only
energetically possible channel from one-photon dissociation in
this energy region, the increase of the overall â indicates that
the contribution of channel 2a, which has a â value g 0, starts
around ET ∼ 63 kcal/mol and increases to the maximum near
43 kcal/mol, while the contribution of the two-photon channel
decreases. In the ET region of 43-70.4 kcal/mol, P(ET) )
2
OH(X2Π) could be readily excited to a repulsive Σ- state by
another 193.3-nm photon via a perpendicular transition and
promptly dissociate into H + O(3P) with a â value of -1.37
The observed strong anisotropy (â ≈ -0.9) of the two-photon
component is in agreement with this mechanism. Furthermore,
the shape of the P2hν(ET) distribution (peaking at ∼63 kcal/
mol) is consistent with the H + O(3P) product translational
energy release from 193.3-nm photodissociation of OH(X2Π)
radicals that are produced largely in vibrational state V ) 0 and
1 and with â ≈ 1.3 in the photolysis of HONO.26,27 Although
conversion of the TOF spectrum into P(ET) is based on NO2 as
the counterfragment, it can essentially apply for the two-photon
component (with O atom as the countermass), since the H-atom
mass is much smaller than that of O and NO2 and the difference
in conversion is negligible.
The P(ET) distribution and â parameter of 193.3-nm photo-
dissociation of HONO, with the two-photon contribution
removed, are shown in Figure 7. Note that for the H +
NO2(X˜ 2A1) channel, â is chosen to be 0 (i.e., the lower limit).
For ET e 70.4 kcal/mol (maximum available energy), several
distinct regions can be identified in the TOF spectra, the P(ET)
distributions, and in particular, the energy-dependent â param-
eter (Figures 4, 5, and 7). It seems that, besides the fast two-
photon channel, several product channels are present, most likely
corresponding to different electronic states of NO2. The first
one starts at ∼15.5 µs and continues until the pronounced break
at ∼20 µs in the TOF spectra. The H + NO2(X˜ 2A1) channel is
unambiguously identified in this region (ET from 70.4 to 43
kcal/mol). The second region starts at ∼20 µs (ET ∼ 43 kcal/
mol) with a different slope from that of the first region in the
TOF spectra, a resolved vibrational structure, and a significant
P
2hV(ET) + P1hV(ET), and the overall â is the weighted sum,
â(ET) ) ø2hV(ET)â2hV + ø1hV(ET) â1hV, with øihV(ET) ) PihV(ET)/

1470 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Amaral et al.
TABLE 1: Properties of the H + NO2 Product Channels
relative
branching
ratios
Eavail
(kcal/mol)
a
a
a
â
fT
fele fvib frot.
overall
70.4
70.4
42.6
30.0
27.9
0.30 0.44
0.63
0.74
0.44
0.47
0.26
0.37
0.22 0.04
0.56
H + NO2(X˜ 2A1)
H + NO2(A˜ 2B2)
H + NO2(B˜2B1)
H + NO2(C˜2A2)
∼ 0
0.13
0.21
-0.4
-0.2
-0.2
0.66
0.53
a Fraction of energy is based on the Eavail of each product channel.
NO2(C˜2A2). Below ET < 22 kcal/mol, â stays effectively at an
average value of -0.2, which implies that the branching ratio
of the H + NO2(A˜ 2B2) channel becomes negligible. Interest-
ingly, the â value always changes drastically near the onsets of
new product channels: at ∼43 kcal/mol for NO2(A˜ 2B2) and at
∼30 kcal/mol for NO2(B˜2B1) or NO2(C˜2A2). Assuming a
constant â value for each individual channel and using â = 0
for NO2(X˜ 2A1), -0.4 (the observed minimum) for NO2(A˜ 2B2),
and -0.2 for NO2(B˜2B1) and/or NO2(C˜2A2), the relative
contribution, Pi(ET), of each of these electronic states can be
deconvoluted by using eqs 7 and 8, and they are shown in the
three regions in Figure 7a. Although the details at the edges of
each of these distributions are extracted with some uncertainty,
the global features of these Pi(ET)’s should be reliable.
Figure 7. (a) One-photon P1hV(ET) (from Figure 5b) and its decon-
volution into three H + NO2 product channels. The onsets of the H +
NO2 channels are marked. The Pi(ET)’s in regions I, II, and III
correspond to the H + NO2(X˜ 2A1), H + NO2(A˜ 2B2), and H + NO2-
(B˜2B1) and/or H + NO2(C˜2A2) channels; (b) â parameter for one-photon
dissociation of HONO (derived by removing the two-photon component
from Figure 5d). See the text for details.
change of â. Since the threshold of the NO2(A˜ 2B2) product lies
at ET ) 42.6 kcal/mol, the second region likely corresponds to
the H + NO2(A˜ 2B2) channel. The third region begins with an
abrupt change of slope in the TOF spectra and a change of â at
∼25 µs (ET ∼ 30 kcal/mol), around which the B˜2B1 and/or C˜2A2
states of NO2 start.
The observed P(ET) distribution is thus a sum of these
dissociation pathways, and the energy-dependent â is due to
these multiple channels that have different â parameters and
different energy-dependent branching ratios.36 Accordingly, the
overall P(ET) distribution and â can be expressed as36
The overall P(ET) in Figure 7a indicates that a large amount
of available energy is partitioned into internal energy of the
NO2 product (electronic, vibrational, and rotational). The overall
average translational energy ET ) 0.30Eavail. At least three
electronic states of NO2 are produced, with the higher B˜ and/or
C˜ states mostly populated and an inverted electronic energy
distribution. The vibrational state population in each of these
NO2 electronic states is highly inverted as well. P(ET)’s are
structureless in regions I and III, while in region II some
vibrational structure is resolved. The relative branching ratios
of these product channels are estimated on the basis of the areas
of the Pi(ET) distributions in Figure 7a: X˜ 2A1:A˜ 2B2: B˜2B1/C˜2A2
≈ 0.13:0.21:0.66. Note that P(ET) [obtained from Pm(ET)] is
the angle-integrated distribution and thus has been corrected
for the different angular distributions of the product channels.
The estimated errors of the branching ratios are in the order of
20% of the listed values. The summed results for these product
channels are listed in Table 1.
n
P(E ) ) P (E )
(7)
(8)
∑
T
i
T
i
n
â(E ) ) ø (E )â
i
∑
T
i
T
i
where Pi(ET) and âi are for the ith channel and øi ) Pi(ET)/
Discussion
P(ET) is the branching fraction of the ith channel (with
n
1. Electronically Excited States and Transition Moments
of HONO. The 193.3-nm photoexcitation involves the second
absorption band of HONO, which is due to the B˜1A′ r X˜ 1A′
transition via the 2a′′(πnb,O) f 3a′′[π*(NdO)] excitation.5a,7,9
The transition dipole moment µ of this B˜ r X˜ transition is
therefore in the molecular plane. The angle between µ of the B˜
r X˜ transition and the dissociating N-O bond of HONO was
calculated to be ∼39° (in the ground-state geometry of HONO,
see Figure 8).27 This angle has been confirmed by an experi-
mental value in the range of 29-35° based on the vector
correlations and Doppler profiles of OH LIF for the OH + NO
channel by Vasudev et al.26,27 For the H + NO2 channels, the
angle between the in-plane µ and the O-H bond direction is
∼63° based on the theoretical calculation (Figure 8). In the limit
of prompt dissociation and axial recoil, the O-H bond direction
is taken as the H atom recoil direction v, and the anisotropy
parameter is expressed as â ) 2P2(cos R), where R is the µ-v
angle.35 Thus, with a theoretical value R ) 63°, a â value of
∑øi(ET))1). Equations 7 and 8 can be readily derived from eq
i
4, with the assumption that the overall differential cross-sections
are the simple sum (with no coherent interference) of those of
the individual channels. The H + NO2(X˜ 2A1) channel starts from
ET ) 70.4 kcal/mol, with a â ≈ 0 (the lower limit). Starting
from the threshold region of H + NO2(A˜ 2B2) (ET ) 42.6 kcal/
mol), the overall â decreases from ∼0 and reaches the minimum
of -0.4 at ET ∼ 30 kcal/mol (Figure 7b). This indicates that
contribution of NO2(A˜ 2B2) increases to the maximum, while
that of NO2(X˜ 2A1) decreases to the minimum and becomes
insignificant below 30 kcal/mol. As the H + NO2(B˜2B1) channel
opens at ET ) 30.0 kcal/mol and the H + NO2(C˜2A2) channel
at 27.9 kcal/mol, the “turn-around” or increase of the â
parameter from ET < 30 kcal/mol is most likely caused by the
increasing contribution of the H + NO2(B˜2B1) and/or NO2(C˜2A2)
channels, rather than the H + NO2(X˜ 2A1) channel. It is difficult
to separate the individual contributions of NO2(B˜2B1) and

H + NO2 Channels in the Photodissociation of HONO
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1471
state surface of HONO(X˜ 1A′), which has the same symmetry
for vibronic couplings, and then dissociate to H + NO2(X˜ 2A1).
Although the â g 0 might be consistent with unimolecular
decomposition of HONO(X˜ 1A′) into H + NO2(X˜ 2A1) in a time
scale longer than a rotational period, the large product trans-
lational energy release contradicts this mechanism. An alterna-
tive is that the excited B˜1A′ HONO could undergo intersystem
crossing to the b˜3A′ state, which then dissociates into the H +
NO2(X˜ 2A1) products. Ab initio calculation has identified a
dissociation barrier along the H-ONO coordinate on the b˜3A′
surface in the planar fragmentation geometry,9 and thus the
HONO dissociation, with enough energy provided by the 193.3-
nm excitation, could have a repulsive translational energy release
after passing this barrier. The â value of g0 for the H +
NO2(X˜ 2A1) channel might then be due to a change of the initial
geometry of HONO in going from the B˜1A′ state to the b˜3A′
state and the subsequent H-atom departure on the triplet surface
via an exit-channel geometry with a µ-v angle of ∼55°.
Figure 8. Equilibrium geometry of HONO(X˜ 1A′) (based on ref 38),
NO2(X˜ 2A1), NO2(A˜ 2B2), NO2(B˜2B1), and NO2(C˜2A2) (based on refs 17
and 39). Bond length is in angstroms, and bond angle is in degrees.
The transition dipole moment µ of the B˜1A′ r X˜ 1A′ transition in HONO
is indicated.
The large vibrational excitation in NO2 could be rationalized
by the significant change in geometry between the HONO parent
and the NO2(X˜ 2A1) product. Specifically, the ONO bond angle
changes from the equilibrium value of 111° in the ground-state
HONO38 to 134° in NO2(X˜ 2A1).17,39 Therefore, bending vibra-
tional excitation in NO2 is expected in going from the parent
molecule [which determines the geometric structure of the
Franck-Condon (FC) region on the excited-state HONO(B˜1A′)]
to the product in a fast dissociation. With only the bending
excitation considered, the peak internal energy of NO2(X˜ 2A1)
corresponds to V2 (bending vibration) ∼12 and an ONO bond
angle of ∼107° at the classic turning point on the bending
potential energy curve of NO2(X˜ 2A1).17a,40 This result implies
that the excited HONO parent has an effective ONO angle of
107°, consistent with the initial ONO bond angle of 111° in
the FC region. The region where the excited states of HONO
couple, as well as the exit channel, may differ from the FC
configuration and require some change in the ONO angle.
Finally, other vibrational modes of NO2 could also be populated,
which, along with a high rotational excitation, may account for
the lack of resolved vibrational structure in the spectrum.
-0.38 is then derived. The experimental â values of -0.4 and
-0.2 for region II and region III are consistent with the
expectations, as discussed below. Note that if the HONO
geometry and the R angle at the exit channel differ from those
of the initial configuration, the â value will change accordingly,
and in addition, when the dissociation process occurs at a longer
time scale, â will become close to 0, giving a more isotropic
angular distribution. For cis-HONO, assuming the same angle
between µ and the O-N bond, a â of ∼0.4 would be predicted
for a prompt H + NO2 channel, which is clearly not observed
in the experiment.
The third singlet excited state (C˜1A′′) may be involved in
the second absorption band as well. The transition dipole
moment µ of the C˜ r X˜ transition is perpendicular to the
molecular plane, and the â parameter should be -1 for a prompt
dissociation.
2. H + NO2(X˜ 2A1) Channel. The first region of the P(ET)
distribution in Figure 7a corresponds to production of the H +
NO2(X˜ 2A1) fragments. Since H + NO2(X˜ 2A1) is the only
possible product channel in the range of ET ) 43-70.4 kcal/
mol, this assignment is unambiguous. The shape of this
structureless peak (e.g., the slope around 43 kcal/mol in Figures
5a and 7a) does not appear to continue into the second region
(the A˜ state of NO2, with a resolved vibrational structure) or
into the third region (the B˜ and/or C˜ state of NO2). More
importantly, the energy-dependent â parameter indicates that
the NO2(X˜ 2A1) product has a different â from the NO2(A˜ 2B2)
and NO2(B˜2B1) products (which are in the lower ET region).
Therefore, the contribution of the X˜ 2A1 state (peak I, Figure
7a) is concentrated in the region of 30-60 kcal/mol (i.e., not
extended into or insignificant in the region below 30 kcal/mol
where the B˜ and/or C˜ states start). Translational energy release
of the NO2(X˜ 2A1) product channel is large (with fT ≈ 0.63).
Vibrational population of NO2(X˜ 2A1), although not resolved,
is clearly inverted. Conversely, the â value of g 0 for the H +
NO2(X˜ 2A1) channel seems to indicate a nearly isotropic product
angular distribution.
It is well-known that the X˜ 2A1 and A˜ 2B2 states of NO2 are
coupled by vibronic interactions via the asymmetric stretching
mode, which results in a conical intersection of the X˜ 2A1 and
A˜ 2B2 surfaces.14,16,17,40-46 The vibronic coupling starts near the
minimum of the A˜ 2B2 surface (9726 cm-1, corresponding to
ET ) 42.6 kcal/mol in this experiment), and increases gradually
and selectively with energy up to 12 500 cm-1, depending on
the near resonance of the X˜ 2A1 and A˜ 2B2 zero-order levels and
the coupling matrix element.16,41,42 The vibronic coupling further
increases above 12 500 cm-1, but some regularities or clumps
of the strongly mixed vibronic bands, which are mainly caused
by A˜ 2B2 bending progression, are still present.16,42 Above 16 000
cm-1, the nonadiabatic coupling fully mixes the two electronic
states, generating a chaotic spectrum.42,43 In this experiment,
another possibility of the production of NO2(X˜ 2A1) is via the
vibronic coupling with the excited-state NO2(A˜ 2B2) fragment,
which, based on its adiabatic correlation with HONO(B˜1A′) in
the coplanar geometry, could be readily produced (discussed
below). The coupling of the X˜ 2A1 and A˜ 2B2 states of NO2 could
be developed (in subpicosecond time scale)44 before the H atom
completely departs, and the H atoms can therefore sample the
mixed X˜ 2A1 character in the H + NO2(A˜ 2B2) channel in final
state interactions. However, since the vibronic coupling starts
only above ∼9700 cm-1, this plausible pathway requires an
extensive energy transfer from the internal modes of the ONO
Since the H + NO2(X˜ 2A1) channel is not correlated adiabati-
cally with HONO(B˜1A′) in either coplanar or nonplanar
geometries, production of the ground-state NO2(X˜ 2A1) requires
nonadiabatic processes in HONO. One possibility is that the
B˜1A′ HONO could undergo internal conversion to the ground-

1472 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Amaral et al.
moiety to H-atom translation. Furthermore, since the vibronic
coupling is weak in this low-energy region (with only ∼10%
of the highly excited vibrational levels in the X˜ 2A1 state coupled
with the A˜ 2B2 state and with small mixing coefficients),16a,42
the fraction of the X˜ 2A1 character due to coupling with A˜ 2B2
may not be significant in the H + NO2(X˜ 2A1) channel.
3. H + NO2(A˜ 2B2) Channel. As discussed previously, the
second region corresponds to the formation of H + NO2(A˜ 2B2),
which correlates adiabatically in the planar geometry with the
HONO B˜1A′ excited state.9,14 The H + NO2(A˜ 2B2) channel has
a pronounced anisotropy in the H-atom angular distribution (with
â ) -0.4). This experimental â parameter is in good agreement
with the theoretical value of -0.38, which is based on the
theoretical µ-v angle R ) 63° and an assumption of prompt
dissociation.27 Note that the â ) -0.4 is an upper limit due to
the deconvolution procedures. The observed anisotropy favors
a prompt dissociation mechanism in which HONO(B˜1A′)
dissociates directly into the H + NO2(A˜ 2B2) fragments in a short
time less than a rotational period. The translational energy
release is large and repulsive [fT ) 0.74, based on Eavail of the
H + NO2(A˜ 2B2) channel]. A dissociation barrier along the
H-ONO coordinate on the B˜1A′ surface (obtained in a planar
geometry and without full optimization of the geometric
structure along the reaction path) has been identified by the
theoretical calculation.9 The repulsive energy release is therefore
consistent with this repulsive barrier along the O-H dissociation
coordinate and with the large available energy from the 193.3-
nm photon (which readily overpasses the barrier). In addition,
the specific product vibrational state population (discussed
below) is also in agreement with the prompt photodissociation
mechanism. On the other hand, the B˜1A′ state might have a
nonplanar equilibrium geometry as suggested by Vasudev et
al., due to the removal of the out-of-plane pπ electron on the
OH oxygen atom and reduction of the lone pair repulsion.26,27
Upon vertical excitation to the B˜ state, the OH moiety would
undergo an out-of-plane motion toward the equilibrium con-
figuration, generating significant excitation in the out-of-plane
torsion ν6.26,27 In the nonplanar fragmentation geometries, the
H + NO2(A˜ 2B2) channel does not correlate with HONO(B˜1A′),
and thus nonadiabatic process such as hopping to the A˜ 1A′′ or
b˜3A′ surfaces is needed. However, the planar geometry may
still be accessed periodically when the H atom oscillates in and
out of the molecular plane, and the direct production of H +
NO2(A˜ 2B2) channel can still occur under this condition. Indeed,
our experimental results favor a direct dissociation mechanism
for the H + NO2(A˜ 2B2) channel.
Figure 9. NO2(A˜ 2B2) internal energy distribution, P(Eint), from HONO
photodissociation at 193.3 nm. This is obtained by conversion of the
NO2(A˜ 2B2) peak in Figure 7a using eq 5 and a shift by the threshold
energy of NO2(A˜ 2B2). The Gaussian peak fits of the NO2 bending
progression are plotted. Circles are experimental data, the solid line is
the fitted spectrum, and dashed lines are Gaussian peaks whose widths
include rotational envelopes and laser line width. The NO2 bending
progression is labeled in the figure.
can be treated within isolated polyads.16,17,41,42 Therefore, the
vibrational structures of A˜ 2B2 near the conical intersection are
still regular and can be identified as polyads.16,17b,41,42 In
addition, A˜ 2B2 bending vibration has been shown to be the
dominant character of the first couple of polyads (at a spacing
of ∼740 cm-1) in the low-energy region.16,17b,41,42 In this
experiment, the resolution is in the order of 150 cm-1 (mainly
due to the excitation laser line width), which does not allow
resolution of the X˜ 2A1-A˜ 2B2 vibronic couplings with off-
diagonal matrix elements in the order of 50 cm-1 16b,45,46
.
Therefore, a “zero-order” picture, in which the vibronic interac-
tions of the X˜ 2A1 and A˜ 2B2 states are negligible, would be
appropriate for the low-resolution P(Eint) distribution in the low-
energy region near the origin of the A˜ 2B2 state. This is to say
that, although the strong X˜ 2A1-A˜ 2B2 vibronic couplings and
the vibrational couplings in the A˜ 2B2 state cause complex
vibronic structures in the polyads, the low-resolution P(Eint)
distribution in this experiment is not sensitive to these couplings
that have small splittings and can only be detected with higher
energy resolution.
To qualitatively model the vibrational population, P(Eint) in
Figure 9 is deconvoluted by fitting with several Gaussian peaks,
whose positions, heights, and widths are adjusted. Peak width
is chosen to be nearly the same for all the peaks (∼600 cm-1).
The peak width reflects the excimer photolysis laser line width,
the rotational excitation in the NO2 fragment, and the envelope
of the perturbed vibrational levels. These peaks are evenly
spaced (except the last three) with a separation of 735 ( 50
cm-1, which is consistent with a bending progression based
on the literature value of the bending frequency of NO2-
(A˜ 2B2).16,17,41,42 The bending vibration of NO2(A˜ 2B2) is highly
excited, with a maximum population of ν2 ) 6 and an average
internal energy of 11 kcal/mol. It is important to note that P(Eint)
of NO2(A˜ 2B2) is extracted from the P(ET) distribution with some
uncertainty, and the fitted vibrational distribution is qualitative
and may not be unique. For example, possible ν1 and ν3 stretch
excitations, which are due to the change of N-O bond length
during the dissociation, may be buried in the P(Eint) and cannot
be ruled out, and the Fermi resonances (with ν1 = 2ν2 = ν3)
can further complicate the vibrational structures. Nevertheless,
the main conclusion of a predominant bending progression that
peaks around ν2 ) 6 in the NO2(A˜ 2B2) product is robust.
The equilibrium ONO bond angle is 102° in the excited-state
NO2(A˜ 2B2)17,39 and 111° in the ground-state HONO.38 On the
The internal (vibrational and rotational) energy distribution
of the NO2(A˜ 2B2) product, P(Eint), is shown in Figure 9, which
is obtained by using eq 5 and a shift of the electronic energy of
NO2(A˜ 2B2). This distribution reveals some vibrational structure,
which is similar to the bending progression (with an oscillatory
structure at a spacing of ∼740 cm-1) of the A˜ 2B2 electronic
state near the same energy region in the near-IR absorption
17b,41
spectrum of NO2
trum of NO2
and the photoelectron detachment spec-
.
- 46,47 As discussed before, the vibronic structures
of the A˜ 2B2 state are deeply influenced by the conical intersec-
tion between the X˜ 2A1 and A˜ 2B2 states and by Fermi resonances
in the A˜ 2B2 state. The Fermi resonances, due to the approximate
2:1:2 ratio in the three vibrational frequencies in the A˜ 2B2 state
(symmetric stretch ν1 ≈ 1484 cm-1, bending ν2 ≈ 740 cm-1
,
and antisymmetric stretch ν3 ≈1500 cm-1), cause the polyad
structure.16,17,41,42 In the low-energy region up to 12 500 cm-1
,
the vibronic interactions between X˜ 2A1 and A˜ 2B2 are weaker
than the average vibronic spacing, and thus are localized and

H + NO2 Channels in the Photodissociation of HONO
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1473
∼9° from the 111° at the FC region, and this enlarged ONO
bond angle in HONO causes a more extensive bending
excitation in the NO2(A˜ 2B2) (whose equilibrium angle is 102°)
when the H atom departs. It is essential to point out that the
FC modeling is to give a guideline for understanding the bending
population, and in addition, the de-convolution and fitting of
P(Eint) are affected by the associated uncertainties. Therefore,
the ∼9° increase in the ONO angle has to be treated as a
qualitative value. Nevertheless, the picture of a significant
change of the ONO angle in the course of O-H bond
dissociation is reliable. This conclusion is supported by examin-
ing the bending PES of NO2(A˜ 2B2):17a,40 an O-N-O angle of
∼120° is required to produce a bending progression that peaks
at V2 ) 6. An NO2(A˜ 2B2) bending progression peaking at V2 )
4 in the photoelectron detachment spectrum of NO2-, in going
Figure 10. Franck-Condon modeling of the bending vibration
excitation in NO2(A˜ 2B2). The shaded bars are the experimental bending
vibration population, and the unfilled bars are for the Franck-Condon
modeling. The total area of each of the two distributions is normalized
to unity. See text for more details.
-
from an equilibrium angle of 117.5° in NO2 to 102° in
NO2(A˜ 2B2), also supports our Franck-Condon modeling and
the conclusion of a significant change of O-N-O angle.47 It
seems that the effective µ-v angle R (between the O-H bond
and µ) does not change significantly in the course of the
dissociation, since the observed â parameter (-0.4) is in good
agreement with the theoretical value (-0.38) based on the initial
geometry of the FC region.26,27 Finally, a decrease of the
O-N-O angle in the excited-state HONO is also possible in
the FC modeling, with θeq(O-N-O) ∼ 86° and a decrease of
∼25° from 111° in the FC region. The above discussion of the
NO2(A˜ 2B2) vibrational distribution can be applied qualitatively
as well. However, due to the modest O-N-O angle dependence
of the 2a′′ and 3a′′ orbitals of HONO (in the 2a′′ f 3a′′
transition),9,48 a large decrease of the O-N-O angle is unlikely
(at least in the FC region).
basis of the Franck-Condon principle, upon 193.3-nm photo-
excitation, HONO on the B˜1A′ surface is prepared with the same
geometry as in the ground state, which is the starting point of
the photodissociation of the exited-state HONO(B˜1A′). Vibra-
tional excitation in the NO2 bending mode is therefore expected
from the change of the ONO angle in going from the parent
molecule in the excited state to the product in a prompt
dissociation. However, the change of the ONO angle between
the initial geometry of the FC region on the B˜1A′ surface
(essentially the geometry of the equilibrium ground state) and
NO2(A˜ 2B2) is not large enough to account for the extensive and
inverted ONO bending progression. Additional change in the
ONO bond angle while the O-H bond is dissociating is required
to generate the observed bending progression. A simple one-
dimensional Franck-Condon calculation (which considers only
the NO2 bending oscillator) has been carried out to model the
observed bending progression. In this model, the ONO bending
wave function of the excited-state HONO is projected onto the
basis set of the bending vibration of NO2(A˜ 2B2), and under the
assumption of a prompt and sudden dissociation, the NO2(A˜ 2B2)
bending vibrational distribution can be obtained as the overlap
integrals of the ONO bending vibrational wave function of the
excited-state HONO and those of NO2(A˜ 2B2). The ONO bending
modes of the excited-state HONO and NO2(A˜ 2B2) are repre-
sented by harmonic oscillators in the simulation. For the
NO2(A˜ 2B2) bending, literature values of ν2 ∼ 740 cm-1 and
θeq(NO2) ) 102° are used. The ONO bond angle and bending
frequency of the excited-state HONO configuration at which
the H atom departs (or at the transition state) are assumed to be
the equilibrium angle, θeq(O-N-O), and the frequency, ν, of
the O-N-O bending harmonic oscillator potential in HONO.
Note that, on the H-ONO part of the excited-state PES of
HONO, the ONO moiety is bound. Both θeq(O-N-O) and ν
of the excited-state HONO are optimized [with θeq(O-N-O)
) 120° ( 1° and ν ≈ 740 ( 100 cm-1] to achieve the best fit
of the experimental NO2 bending progression (shown in Figure
10). Despite the crudeness of the simulation, the simple FC
modeling indicates qualitatively that, starting from the FC region
and determined by the topology of the excited-state PES, there
is a potential gradient toward opening of the O-N-O bond
angle, besides the force toward the O-H dissociation. This
angular gradient creates a significant bending motion in the
excited HONO, which is then carried into the NO2 product while
the excited HONO dissociates on the excited surface along the
O-H coordinate. Specifically, the ONO angle is increased by
4. H + NO2(B˜2B1) and/or NO2(C˜ 2A2) Channels. The
thresholds of the H + NO2(B˜2B1) and H + NO2(C˜2A2) channels
are ET ) 30.0 and 27.9 kcal/mol, respectively.17 The third
component in the P(ET) distribution in Figure 7a, starting in
this region, should correspond to these two channels. This third
component peaks at ET ∼ 10 kcal/mol and has an inverted
vibrational population in NO2(B˜2B1)/NO2(C˜2A2), although little
vibrational structure is resolved (partially due to the poor S/N
ratio). This high internal energy region corresponds to the well-
known NO2 vibrational chaos, which appears at internal energies
above 45 kcal/mol (or ET < 24 kcal/mol) and is caused by the
strong X˜ 2A1-A˜ 2B2 vibronic coupling and the Fermi resonances
in the A˜ 2B2 state.16,42,43 In addition, the X˜ 2A1 and C˜2B1 states
and the A˜ 2B2 and C˜2A2 states are coupled by Renner-Teller
effect, respectively, and the B˜2B1 and C˜2A2 states could be
coupled by vibronic coupling as well.16,17 Although these
couplings are in the free NO2 molecules, it is possible that they
start to occur (e.g., A˜ 2B2 f X˜ 2A1 within 100 fs),44 while the
H-atom fragment is leaving but still interacting with the NO2
moiety during the dissociation. The congestion of the P(ET)
distribution in this region could be due to the strong mixings
of the electronic states involved. A slightly anisotropic angular
distribution in this region is observed with an average â ≈ -0.2.
As the NO2(B˜2B1) correlates adiabatically in the planar
geometry with the A˜ 1A′′ state and a˜3A′′ state of HONO,
production of the NO2(B˜2B1) from the initially excited B˜1A′
HONO requires nonadiabatic processes such as internal conver-
sion from the B˜1A′ surface to A˜ 1A′′ or intersystem crossing to
a˜3A′′. The out-of-plane torsion excitation [V6(a′′)] in the B˜ r
X˜ transition, as suggested by the previous study,26,27 could
facilitate the internal conversion from B˜1A′ to A˜ 1A′′ with an
A′′ vibronic symmetry. Alternatively, in the nonplanar geom-
etries, the NO2(B˜2B1) product correlates adiabatically with the

1474 J. Phys. Chem. A, Vol. 105, No. 9, 2001
Amaral et al.
B˜1A′ HONO. Consequently, direct production of the H +
NO2(B˜2B1) channel could explain its dominant branching ratio.
The equilibrium ONO angle in NO2(B˜2B1) is 180°, much larger
than that in the HONO; thus a large amount of NO2 bending
excitation is expected. The slightly less-negative â value of -0.2
could reflect the lengthened lifetime of the coplanar dissociation
process in going from the B˜1A′ to A˜ 1A′′ or a˜3A′′ and then to
the products. Alternatively, it could also indicate a different µ-v
angle R (∼59°) when the H atom departs in the H + NO2(B˜2B1)
channel in the nonplanar dissociation.
ment.9,14,17,48 This could reduce the rate of the adiabatic H +
NO2(A˜ 2B2) channel in the planar dissociation, and thus the
nonadiabatic channels become competitive. A close examination
of orbital correlations indicates that the excited orbitals of
HONO(B˜1A′), 2a′′ f 3a′′, correlate with 1a2 f 2b1 orbitals in
a higher excited-state NO2(22B2) [...(1a2)1(4b2)2(6a1)1(2b1)1],
while the 4b2 f 6a1 orbitals in NO2(A˜ 2B2) correlate with a
higher state of HONO.9,14,17,48 Since both correlations have A′
symmetry in the planar geometry, an avoided crossing will
result. This avoided crossing between the Franck-Condon
(molecular) region and the asymptotic (product) region would
then cause the apparent barrier on the adiabatic potential surface
connecting HONO(B˜1A′) and H + NO2(A˜ 2B2).9,33 The barriers
on other excited-state surfaces of HONO in the planar geometry
could arise for the same reason.
At 193.3 nm, HONO could be excited to the C˜1A′′ state as
well, which correlates directly to NO2(C˜2A2). A vertical
transition to the C˜1A′′ state at 188 nm, as opposed to 213 nm
to the B˜1A′ state, has been estimated by Larrieu et al.9
A
considerable barrier for the H-O cleavage on the HONO(C˜1A′′)
surface is also indicated by the calculation.9 As the transition
moment of the C˜ r X˜ transition is perpendicular to the HONO
molecular plane and thus perpendicular to the recoiling velocity
of the H fragment, a negative â value of -0.2 is also consistent
with dissociation from the C˜ state. However, the lifetime should
be a significant fraction of the rotational period, as the â is
smaller in value than the theoretical value of -1 for a prompt
dissociation. Further experiment at a slightly longer wavelength
where only the B˜ state is excited could separate the contributions
of these two possible dissociation pathways.
Conclusions
The H-atom product channels in the photodissociation of
HONO at 193.3 nm via the B˜1A′ state have been observed for
the first time. Product CM translational energy distribution and
energy-dependent angular distribution indicate that the NO2
products are formed in at least three different electronic states,
with the fraction of the overall average product CM translational
energy fT being 0.30. The relative branching ratios of the
different NO2 electronic channels are estimated to be: X˜ 2A1:
A˜ 2B2:B˜2B1/C˜2A2 ≈ 0.13:0.21:0.66.
5. Relative Branching Ratios of the H + NO2 and the OH
+ NO Channels. The broad and structureless feature of the
second absorption band of HONO indicates a short excited-
state lifetime. The dissociation lifetime of the lower A˜ 1A′′ state
of HONO into OH + NO has been shown to be in the order of
∼100 fs. The dissociation time scale of HONO(B˜1A′) into OH
+ NO should be comparable, since HONO(B˜1A′) also correlates
adiabatically with the OH + NO channel and, in particular, its
PES curve along the N-O dissociation coordinate is steeper
than the one correlating to HONO(A˜ 1A′′).9 The short HONO-
(B˜1A′) lifetime (in femtoseconds) with respect to OH + NO
dissociation is consistent with the observed anisotropic OH +
NO product angular distribution by Vasudev et al.26,27 The
dissociation lifetimes of the H + NO2 channels are in general
less than or comparable to the rotational period (in the order of
∼ps) based on the H-atom angular distribution. The potential
barriers along the H-O dissociation coordinate on the excited-
state surfaces, as well as the nonadiabatic processes involved
in some H + NO2 channels, may slightly hinder the O-H
dissociation processes. The small signal intensities of the H +
NO2 channels observed in this experiment imply their smaller
branching ratios compared to those of the OH + NO channels.
Among the H + NO2 channels, the H + NO2(A˜ 2B2) channel
(adiabatic in the planar geometry and nonadiabatic in the
nonplanar geometries) is not dominant, while the H + NO2(B˜2B1)
(nonadiabatic in planar and adiabatic in nonplanar) or H +
NO2(C˜2A2) is the most important one, and the nonadiabatic
channel H + NO2(X˜ 2A1) is also significant. The large branching
ratio of H + NO2(B˜2B1) channel could be rationalized by a
nonplanar dissociation mechanism, in which the HONO(B˜1A′)
correlates adiabatically with H + NO2(B˜2B1). In addition,
although the H + NO2(A˜ 2B2) channel correlates adiabatically
with HONO(B˜1A′) in the planar geometry, this channel is
Woodward-Hoffmann symmetry forbidden.33 The individual
molecular orbital symmetry of the NO2 moiety in the excited-
state HONO(B˜1A′) is a′′a′′ (2a′′ f 3a′′, involving two out-of-
plane electrons), but it has to change into the a′a′ symmetry
(4b2 f 6a1, involving two in-plane electrons) in the NO2(A˜ 2B2)
product through a double-substitution electronic rearrange-
The ground-state NO2(X˜ 2A1) could be produced in the
nonadiabatic processes via a triplet state (e.g., b˜3A′) or the X˜ 1A′
state of HONO. The NO2(X˜ 2A1) product has a highly inverted
vibrational population, which is likely due to the change of the
ONO angle in the region of intersystem crossing or internal
conversion between the different HONO electronic states and
subsequent dissociation. The large product translational energy
seems to favor the dissociation process from a triplet state that
has a repulsive dissociation barrier along the O-H coordinate.
The NO2(A˜ 2B2) product shows a large bending vibration
excitation (peaking at V2 ) 6), and a simple Franck-Condon
modeling indicates a significant change of the ONO angle during
dissociation. The observed anisotropy parameter in the product
angular distribution is in agreement with the predicted value
based on the known in-plane transition dipole moment and a
prompt dissociation mechanism, which suggests a short excited
B˜1A′ state lifetime (less than one rotational period) with respect
to the H + NO2(A˜ 2B2) dissociation. A large fraction of the
available energy of this channel is partitioned into product
translation. The H + NO2(A˜ 2B2) channel is produced via a direct
dissociation, which could occur via the coplanar geometry in
which H + NO2(A˜ 2B2) correlated adiabatically with HONO-
(B˜1A′), or from the nonplanar geometries that occasionally
sample the near-planar configurations.
The vibrational population of the NO2(B˜2B1) and/or NO2(C˜2A2)
states is also inverted, and a small anisotropy of the H-atom
product has been observed. The H + NO2(B˜2B1) channel could
be produced by nonadiabatic processes via the A˜ 1A′′ or a˜3 A′′
surfaces of HONO in the coplanar fragmentation geometry. This
channel could also be produced adiabatically from the
HONO(B˜1A′) in the noplanar geometries, which is consistent
with the large branching ratio of this channel. The H +
NO2(C˜2A2) channel might involve the C˜1A′′ state of HONO.
Acknowledgment. This work is supported by National
Science Foundation (Grant CHE-9811400) and partially by a
UC Regents' Faculty Fellowships and Faculty Development
Award, an ACS-PRF Type G Grant, and a Camille and Henry

H + NO2 Channels in the Photodissociation of HONO
J. Phys. Chem. A, Vol. 105, No. 9, 2001 1475
correction) is 1.75 eV for NO2(B˜2B1) and ∼1.84 eV for NO2(C˜2A2).
(18) Vasudev, R.; Zare, R. N.; Dixon, R. N. Chem. Phys. Lett. 1983,
96, 399.
Dreyfus New Faculty Award. We thank Prof. Curt Wittig for
donating the excimer laser and a vacuum chamber used in this
work. We thank the reviewers for their valuable comments.
(19) Vasudev, R.; Zare, R. N.; Dixon, R. N. J. Chem. Phys. 1984, 80,
4863.
(20) Novicki, S. W.; Vasudev, R. J. Chem. Phys. 1991, 95, 7269.
(21) Dixon, R. N.; Rieley, H. Chem. Phys. 1989, 137, 307.
(22) Dixon, R. N.; Rieley, H. J. Chem. Phys. 1989, 91, 2308.
(23) Shan, J. H.; Wategaonkar, S. J.; Vasudev, R. Chem. Phys. Lett.
1989, 158, 317.
(24) Holland, S. M.; Stickland, R. J.; Ashfold, M. N. R.; Newnham,
D.; Mills, I. M. J. Chem. Soc., Faraday Trans. 1991, 87, 346.
(25) Burkholder, J. B.; Mellouki, A.; Talukdar, R.; Ravishankara, A.
R. Int. J. Chem. Kinet. 1992, 24, 711.
(26) Novicki, S. W.; Vasudev, R. Chem. Phys. Lett. 1991, 176, 118.
(27) Wategaonkar, S. J.; Shan, J. H.; Vasudev, R. Chem. Phys. 1989,
139, 283.
(28) (a) Schnieder, L.; Meier, W.; Welge, K. H.; Ashfold, M. N. R.;
Western, C. M. J. Chem. Phys. 1990, 92, 7027. (b) Ashfold, M. N. R.;
Lambert, I. R.; Mordaunt, D. H.; Morley, G. P.; Western, C. M. J. Phys.
Chem. 1992, 96, 2938.
(29) Zhang, J.; Dulligan, M.; Wittig, C. J. Phys. Chem. 1995, 99, 7446.
(30) (a) Zhang, J.; Xu, K.; Amaral, G. Chem. Phys. Lett. 1999, 299,
285. (b) Xu, K.; Amaral, G. Zhang, J. J. Chem. Phys. 1999, 111, 6271.
(31) Ning, C. L.; Pfab, J. J. Phys. Chem. 1997, 101, 6008.
(32) Marshall, G.; Midgley, D. Anal. Chem. 1982, 54, 1490.
(33) Myers, T. L.; Forde, N. R.; Hu, B.; Kitchen, D. C.; Butler, L. J. J.
Chem. Phys. 1997, 107, 5361, and references cited therein.
(34) Okabe, H. Photochemistry of Small Molecules; John Wiley &
Sons: New York, 1978.
(35) (a) Zare, R. N. Mol. Photochem. 1972, 4, 1. (b) Yang, S. C.;
Bersohn, R. J. Chem. Phys. 1974, 61, 4400.
(36) (a) Wang, J. H.; Hsu, Y. T.; Liu, K. J. Phys. Chem. 1997, 101,
6593. (b) Wang, J. H.; Liu, K. J. Chem. Phys. 1998, 109, 7107.
(37) Lee, S. J. Phys. Chem. 1995, 99, 13380.
(38) Finnigan, D. J.; Cox, A. P.; Brittain, A. H. J. Chem. Soc., Faraday
Trans. 2 1972, 68, 548.
(39) Kuchitsu, K. ed. Structure of Free Polyatomic Molecules; Springer-
Verlag: Berlin, 1998; pp 72-73, and references cited therein.
(40) Katagiri, H.; Shigeki, K. J. Chem. Phys. 1993, 99, 8805.
(41) Orphal, J.; Dreher, S.; Voigt, S.; Burrows, J. P.; Jost, R.; Delon,
A. J. Chem. Phys. 1998, 109, 10217.
(42) (a) Leonardi, E.; Petrongolo, C. J. Chem. Phys. 1997, 106, 10066.
(b) Leonardi, E.; Petrongolo, C.; Hirsch, G.; Buenker, R. J. J. Chem. Phys.
1996, 105, 9051.
References and Notes
(1) Finlayson-Pitts, B. J.; Pitts, J. N., Jr. Chemistry of the Upper and
Lower Atmosphere; Academic Press: New York, 2000.
(2) Lammel, G.; Cape, J. N. Chem. Soc. ReV. 1996, 25, 361, and
references cited therein.
(3) Jenkin, M. E.; Cox, R. A.; Williams, D. J. Atmos. EnViron. 1988,
22, 487.
(4) (a) Winer, A. M.; Biermann, H. W. Res. Chem. Intermed. 1994,
20, 423. (b) Calvert, J. G.; Yarwood, G.; Dunker, A. M. Res. Chem.
Intermed. 1994, 20, 463.
(5) (a) Cox, R. A.; Derwent, R. G.; J. Photochem. 1976, 6, 23. (b)
Stockwell, W. R.; Calvert, J. G. J. Photochem. 1978, 8, 193. (c) Bongartz,
A.; Kames. J.; Welter, F.; Schurath, U. J. Phys. Chem. 1991, 95, 1076. (d)
Pagsberg, P.; Bjergbakke, E.; Ratajczak, E.; Sillesen, A.; Chem. Phys. Lett.
1997, 272, 383. (e) Vasudev, R. Geophys. Res. Lett. 1990, 17, 2153.
(6) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr,
J. A.; Rossi, M. J.; Troe, J. J. Phys. Chem. Ref. Data 1997, 26, 1329.
(7) Kenner, R. D.; Rohrer, F.; Stuhl, F. J. Phys. Chem. 1986, 90, 2635.
(8) King, G. W.; Moule, D. Can. J. Chem. 1962, 40, 2057.
(9) Larrieu, C.; Dargelos, A.; Chaillet, M. Chem. Phys. Lett. 1982, 91,
465.
(10) Hennig, S.; Untch, A.; Schinke, R.; Nonella, M.; Huber, R. Chem.
Phys. 1989, 129, 93.
(11) Suter, H. U.; Huber, J. R. Chem. Phys. Lett. 1989, 155, 203.
(12) Schinke, R.; Untch. A.; Suter, H. U.; Huber, J. R. J. Chem. Phys.
1991, 94, 7929.
(13) Cotting, R.; Huber, J. R. J. Chem. Phys. 1996, 104, 6208.
(14) Jackels, C. F.; Davidson, E. R. J. Chem. Phys. 1976, 65, 2941.
The a1 and b2 orbitals are symmetric with respect to reflection in the NO2
molecular (yz) plane and thus have a′ symmetry in the Cs point group, while
the a2 and b1 orbitals have a′′ symmetry. Note that in ref 9 the order and
reduced symmetry of the 2B2 and 2B1 states in the Cs point group were
inverted.
(15) (a) Chase, M. W., Jr. NIST-JANAF Themochemical Tables, 4th ed.;
Journal of Physical ahd Chemical Reference Data, Monograph 9; American
Chemical Society: Washington, DC, 1998. D0(HO-NO) (∆H0° for 1a)
and D0(H-ONO) (∆H0° for 2a) are derived based on the ∆fH° (0 K) values
in the tables. (b) Huber, K. P.; Herzberg, G. Molecular Spectra and
Molecular Structure IV. Constants of Diatomic Molecules; Van Nostrand:
New York, 1979.
(43) Delon, A.; Jost, R. J. Chem. Phys. 1991, 95, 5701.
(44) Santoro, F.; Petrongolo, C. J. Chem. Phys. 1999, 110, 4419.
(45) Lievin, J.; Delon, A.; Jost, R. J. Chem. Phys. 1998, 108, 8931.
(46) Delon, A.; Georges, R.; Kirmse, B.; Jost, R. Faraday Discuss. 1995,
102, 117.
(47) Weaver, A.; Metz, R. B.; Bradforth, S. E.; Neumark, D. M. J. Chem.
Phys. 1989, 90, 2070.
(48) Rabalais, J. W.; McDonald, J. M.; Scherr, V.; McGlynn, S. P.;
Chem. ReV. 1970, 71, 73.
(16) (a) Kirmse, Β.; Delon, A.; Jost, R. J. Chem. Phys. 1998, 108, 6638.
The unperturbed electronic energy T0 of NO2(A˜ 2B2) is 9726 ( 5 cm-1. (b)
Delon, A.; Jost, R. J. Chem. Phys. 1991, 95, 5686. (c) Delon, A.; Jost, R.
J. Chem. Phys. 1999, 110, 4300.
(17) (a) Gillispie, G. D.; Khan, A. U.; Wahl, A. C.; Hosteny, R. P.;
Krauss, M. J. Chem. Phys. 1975, 63, 3425. (b) Gillispie, G. D.; Khan, A.
U. J. Chem. Phys. 1976, 65, 1624. (c) Hsu, D.; Monts, D.; Zare, R. N.
Spectral Atlas of Nitrogen Dioxide; Academic Press: New York, 1978.
The electronic energy T0 (derived from theoretical Te with zero point energy