Vibrational level dependence of lifetime of NO2 in the D?2B2 state

Article abstract of DOI:10.1016/S0009-2614(03)00774-7

The vibrational level dependence of the lifetime of NO₂ D?²B₂ has been investigated by using LIF, hole-burning, and photofragment excitation methods. The predissociation rates for the vibrational levels estimated from the

Full text of DOI:10.1016/S0009-2614(03)00774-7

Chemical Physics Letters 374 (2003) 601–607
www.elsevier.com/locate/cplett
Vibrational level dependence of lifetime of
2
~
NO in the D B state
2
2
1
Kazuhide Tsuji , Masashi Ikeda, Junichi Awamura, Akio Kawai,
Kazuhiko Shibuya *
Department of Chemistry, Graduate School of Science and Engineering, Tokyo Institute of Technology, 2-12-1 Ohokayama,
Meguro-ku, Tokyo 152-8551, Japan
Received 19 February 2003; in final form 6 May 2003
Abstract
The vibrational level dependence of the lifetime of NO
photofragment excitation methods. The predissociation rates for the vibrational levels estimated from the line profiles
~
2
D B has been investigated by using LIF, hole-burning, and
2
2
1
increase sharply with excitation energy even blow the second dissociation limit to form NO and O( D). The occurrence
1
~
2
2
of the dissociation pathway into NO and O( D) via the D B states is confirmed by the observation of the appearance
0
0
2
00
thresholds to form various J -levels of NO( P1=2, v ¼ 0), which accord with the calculated dissociation limits. The
1
dissociation mechanism is discussed based on the energy dependence of predissociation rates and O( D) quantum
yields.
Ó 2003 Elsevier Science B.V. All rights reserved.
1
. Introduction
region near the first dissociation limit to form
2
3
NO( P1=2) and O( P) is regarded as the bench
mark system whether photodissociation occurs in
the statistical nature.
The ultraviolet absorption system of NO2
starting from 249.1 nm shows a much simpler
vibronic structure than the visible one. The rota-
tional structure of this ultraviolet absorption sys-
tem was partially analyzed by Ritchie et al. [1] for
2
Nitrogen dioxide (NO ) is one of fundamental
triatomic molecules and an important component
in the atmospheric and combustion chemistry. The
visible absorption spectrum of NO has long been
the subject of intense investigation, because it
shows extremely complex structure due to the
~
strong vibronic coupling between the X A
~
A B
2
2
1
and
states and the photodissociation in the visible
2
2
the first time. They reported that the upper state
2
has B
2
electronic symmetry. The upper state la-
~
beled D has been characterized by Hallin and
Merer [2] from the high-resolution absorption, and
by our group [3,4] from the dispersed fluorescence.
Detailed rotational analysis of the 249.1 nm band
was reported, and the lifetime was also calculated
*
Corresponding author. Fax: +81-3-5734-2655.
E-mail addresses: ktsuji@chem.titech.ac.jp (K. Tsuji), kshi-
buya@chem.titech.ac.jp (K. Shibuya).
1
Also corresponding author.
0
009-2614/03/$ - see front matter Ó 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0009-2614(03)00774-7

602
K. Tsuji et al. / Chemical Physics Letters 374 (2003) 601–607
~
from the linewidth of the D B
2
~
(0,0,0)–X A
2
2
1
(0,0,0)
pressure of 1 atm, was expanded through a pulsed
nozzle into the vacuum chamber. In hole-burning
experiments, the probe frequency was fixed to a
single rotational line in one of the vibronic bands
transition to be 42 ps [2]. The degree of diffuseness
of the band increases rapidly with increasing en-
ergy. For example, the linewidth of the (0,0,0)
band at 249.1 nm is 0.127 cm , while the rota-
tional structure of the (0,0,2) band at 244.7 nm was
not resolved because of diffuseness. It is interesting
À1
~
2
~
2
of the A B
2
fluorescence light from the A B state was de-
2 1
–X A transition at 593 nm and the
2
~
tected, because the 593 nm band is relatively free
from perturbation due to the strong vibronic
~
to note that the origin of the D B state
2
2
À1
~
coupling between the A B and X A states, and
2
~
2
2 1
(T
0
¼ 40125:849 cm ), is located at energy far
À1
above the first dissociation limit (25 128.57 cm )
À1
then shows a simple rotational structure [16]. In
PHOFEX experiment, photofragment NO was
[
second dissociation limit (40 996.43 cm ) to form
5] and located at energy only 871 cm below the
À1
2
probed by monitoring the LIF signal of the A R–
2
1
2
NO(X P) and O( D).
The photodissociation of NO
X P transition of NO induced by another probe
laser.
~
in the D B
2
2
2
state
has been investigated by several researchers [6–15].
Two groups studies the product state distribution
of the nascent NO for the photodissociation of
3. Results and discussion
À1
2
NO at 248.5 nm corresponding to 40 241 cm
which is slightly lower than the second dissociation
1
limit [7,10]. Shafer et al. [11] reported that O( D)
Fig. 1 shows an LIF spectrum of jet-cooled
~ ~
2
2
2 2 1
NO
called (0,0,0) band, together with the rotational
for the D B (0,0,0)–X A (0,0,0) transition,
fragment had a translational energy distribution
peaking at low energy in the photodissociation of
NO at 205.47 nm. The lifetimes of the vibration-
q
q
assignment. Strong
q
P
0
and
weak P and R branches were observed. Be-
0
R branches and
q
2
1
1
~
2
ally excited levels in the D B state of thermally
2
cause of the nuclear spin statistics, ee and oo ro-
tational levels are allowed to exist in the ground
2
averaged NO were determined by spontaneous
resonance Raman scattering polarization mea-
surements [12]. More recently, strong orbital
2
state of NO . All transitions observed in the LIF
spectrum are assigned to a-type rotational transi-
tions, which accords well with the results reported
by Hallin and Merer [2]. Lineshapes of the peaks
observed in the LIF spectrum are apparently
broader than the linewidth of the laser used (0.25
3
alignment has observed for the fragment O( P
J
) in
2
the photodissociation of NO at 212.8 nm using
ion imaging [13]. Evidently, we have only limited
information on the vibrationally selective photo-
~
dissociation of NO in the D B state, which in-
2
À1
cm ) due to the predissociation in the upper state.
The R
2
2
q
cludes the excited state lifetimes, the direct
1
0
branch is free from the overlap of peaks,
observation of O( D) product, and so on. In the
present study, we spectroscopically investigated
the vibrational level dependence of lifetime of NO
2
2
~
in the D B
jet technique.
2
state in combination with a supersonic
2
. Experimental
We employed three spectroscopic techniques to
measure the absorption band profiles due to
~
the D B
2
0
0
0
~
2
2
(m ; m ; m )–X A
1
2
3
1
(0,0,0) transitions of jet-
, namely LIF, spectral hole-burning
cooled NO
and photofragment excitation (PHOFEX) meth-
2
~
2
~
2
2 1
Fig. 1. LIF spectrum of the (0,0,0) band of the D B
transition of jet-cooled NO
–X A
ods. A 1% NO /Ar premixed gas, kept at a backing
2
2
.

K. Tsuji et al. / Chemical Physics Letters 374 (2003) 601–607
603
À1
and we estimate the Lorentzian width from the
curve fitting operation using the Voigt profile,
which is a convolution of Gaussian and Lorentz-
ian profiles, as a fitting function. We assume the
Lorentzian width of 0.42 cm and Gaussian width
À1
of 0.25 cm . We did not use this spectrum to
determine the linewidth, because the S=N ratio is
much worse than that of Fig. 1 and the peak shape
is broaden by saturation of absorption. We also
measured various hole-burning spectra for the
À1
Gaussian width to be 0.25 cm . The Lorentzian
À1
widths calculated are 0.13, 0.14, and 0.13 cm for
q
q
q
00
R (0), R (2), and R (4), respectively. These
0
monitored levels of N ¼ 0, 2, and 4, and the
0
0
q
q
q
q
q
values are almost the same as the reported value of
À1
R (0), R (2), P (2), R (4), and P (4) lines of
0 0 0 0 0
0
.127 cm [2]. We observed no vibronic bands
the (0,0,0) band were observed. Fig. 2b shows the
hole-burning spectrum for the (0,1,0) band. The
linewidth of the (0,1,0) band is also reproduced
by the Lorentzian profile with the linewidth of
except for the (0,0,0) band in the UV region by use
of the LIF method.
The hole-burning spectrum for the (0,0,0) band
À1
is shown in Fig. 2a. The probe frequency was fixed
q
2.4 cm , which is unnecessary to be corrected by
À1
to the R (0) line of the 593 nm band, which means
the N ¼ 0 rotational level in the ground state was
0
the laser linewidth of 0.25 cm . The fact that the
linewidth of the (0,1,0) level is much broader than
that of the (0,0,0) level is consistent with other
00
q
selectively monitored. The R (0) line of the (0,0,0)
0
~
band of the D B
2
~
–X A
2
~
reports [2,12]. The vibrationally excited D B lev-
els higher than the (0,1,0) level could not been
observed in the hole-burning spectra.
In principle, the hole-burning spectroscopy has
a great advantage for observation of the predis-
sociative states, because it measures the light-in-
duced depletion of population in the ground state.
2
2
1
transition was observed as
a dip signal at the pump frequency of 40 126.8
2
À1
cm as seen in Fig. 2a. The peak shape was
roughly reproduced by the Voigt profile with
The hole-burning spectrum of jet-cooled NO is
2
therefore expected to be equivalent to the rota-
tionally selected absorption spectrum. Neverthe-
less, we could not measure the hole-burning
spectrum for the vibronic levels higher than the
(0,1,0) level. The existence of N
one to observe these levels. N
2
O
4
may prevent
4
O is known to
2
dissociate to form electronically excited NO in the
2
UV photolysis [17–19]. In our hole-burning ex-
periment, we might detect not only the visible
fluorescence of NO induced by the 593 nm probe
2
Ã
laser but also the visible NO chemiluminescence
2
generated by the photodissociation of N
50–220 nm (pump laser). The concentration of
in the 1% NO /Ar mixture of 1 atm total
2 4
O at
2
N
2
O
4
2
pressure is estimated to be 6.2% at room temper-
ature (293 K) from the equilibrium constant. The
2 4
absorption cross section of N O monotonically
increases as the light wavelength becomes shorter
Ã
in the UV region below 250 nm [20]. The NO
chemiluminescence due to the UV photolysis of
2
2 4
N O makes a background noise to the probe
~
2
~
2
Fig. 2. Hole-burning spectra of the D B
2
q
–X A
1
transition of
(0) line of the 593
0
(0) lines of the (0,0,0)
signal in the hole-burning spectrum, which dilutes
the probe depletion fraction in the hole-burning
spectrum and makes the observation of the
2 0
NO . The probe laser transition was the R
q
nm band. Upper and lower traces are R
and (0,1,0) bands, respectively.

604
K. Tsuji et al. / Chemical Physics Letters 374 (2003) 601–607
predissociative levels locating above the (0,1,0)
level difficult.
The PHOFEX spectrum was measured by fix-
ing the probe laser frequency to the band head of
Coon et al. and are summarized in Table 1, to-
gether with the lifetimes calculated from the line-
widths.
At the pump laser energies explored in this
PHOFEX study, two possible dissociation chan-
0
0
2
þ
0
P
–
11 branch (J ¼ 8:5 and 9.5) of NO A R (v ¼ 0)
2 00
3
1
X P1=2(v ¼ 0) transition and scanning the pump
nels are accessible: O( P) and O( D) channels.
Slanger et al. [7] studied the nascent NO vibra-
tional distribution by using an LIF technique for
À1
frequency from 40 900 to 44 000 cm . Four broad
bands and a few quite diffuse bands overlapped
with each other are observed in the PHOFEX
spectrum. Fig. 3 shows an example band assigned
to the (1,0,0) band by Coon et al. [21], the upper
À1
the photodissociation of NO
which is slightly lower than the threshold energy of
2
at 40 241 cm
À1
1
40 996 cm for the O( D) channel. They reported
a strongly inverted population with a maximum at
À1
vibronic level of which locates about 300 cm
00
above the second dissociation limit indicated by an
arrow. The positions of the bands observed in this
experiment are almost the same as reported by
v ¼ 6–8 for the nascent NO. McFarlane et al. [10]
studied essentially the same experiment except for
the detection method of nascent NO, and reported
a bimodal vibrational distribution with peaks at
00
v ¼ 0 and 5. Generally speaking, it is difficult to
distinguish the dissociation channel via the pre-
2
~
2
dissociative D B state from the PHOFEX spectra
2
00
obtained by probing NO(X P1=2, v ¼ 0) fragment
in the photolysis energy range from 40 996 to
À1
1
3
4
2 872 cm , because both O( D) and O( P)
00
2
channels contribute to the NO(X P , v ¼ 0)
1=2
production. The dissociation energy of jet-cooled
2
NO required for the production of NO(X P1=2,
2
00
00
1
00
v ¼ 0, J ) and O( D), Elimit(J NO), is calculated as
the sum of the second dissociation limit and the
rotational energies of nascent NO. If NO
2
in the
00
~
2
2
D B state dissociates to form NO(X P, v ¼ 0,
J ) and O( D), the appearance energy should
accord with Elimit(J
obtained by probing NO(X P , v ¼ 0, J ).
2
00
1
~
D B
2
Fig. 3. PHOFEX spectrum of the (1,0,0) band of NO
2
2
2
–
00
~
X A
profile simulated with FWHM of 85.9 cm
) in the PHOFEX spectrum
2
1
transition (dark gray trace). Black trace is the Lorentzian
À1
NO
00
00
.
1=2
Table 1
~
D B
2
~
2 1
–X A transition together with the corresponding lifetimes of the upper levels
2
Band positions and their assignments of NO
2
Band position
À1
Separation
À1
Assignment
Lifetime (this work)
(ps)
Lifetime (literature data)
(ps)
0 0
(m ; m ; m )
1 2
0
(
cm
)
(cm
)
3
a
d
4
4
4
4
4
4
4
0 126.02
0 651.4
0
(0,0,0)
(0,1,0)
(0,0,2)
(1,0,0)
(1,1,0)
(1,0,2)
(2,0,0)
41 Æ 1.6
42 Æ 5
b
e
525.4
718
4.8 Æ 2.2
2.5 Æ 0.5
e
e
0 844
0.155 Æ 0.015
c
e
1 287
1 814
2 116
2 495
1161
0.062 Æ 0.006
0.075 Æ 0.01
c
e
1688
1990
2369
0.1 Æ 0.01
0.125 Æ 0.015
c
0.034 Æ 0.003
c
0.077 Æ 0.008
a
LIF (this work).
b
c
Hole-burning (this work).
PHOFEX (this work).
Absorption [2].
d
e
Resonance Raman [12].

K. Tsuji et al. / Chemical Physics Letters 374 (2003) 601–607
605
Fig. 4 shows the PHOFEX spectra for the
1,0,0) band obtained by probing various rota-
tional levels of the photodissociation product of
with their result obtained from the reaction prod-
uct analysis. They also suggested that the forma-
(
3
tion of O( P) is also competitive above the second
dissociation limit [6]. There is no indication that
2
00
00
NO(X P1=2, v ¼ 0, J ). The shape of the PHO-
3
FEX spectrum strongly depends on the rotational
quantum number J . In each spectrum, one can
determine the appearance energy to observe a
the O( P) channel contributes to the PHOFEX
signal, which implies that nascent NO as a counter
00
3
00
product of O( P) is not populated in the v ¼ 0
2
00
00
2
specific rotational level of NO(X P1 , v ¼ 0, J ).
=2
level of the X P state. Bigio and Grant [22,23]
1=2
The appearance energies are determined to be
À1
studied the two-photon dissociation of NO
the second dissociation limit, where the probable
2
near
00
4
and 13.5, respectively. These values agree well with
1 237, 41 280, and 41 323 cm for J ¼ 11:5, 12.5,
~
1
2
~
2
~
2
principal pathway is X A
1
! A B
2
! D B
2
[9].
00
3
E
limit(J NO), which are indicated by bold arrows in
They mentioned the O( D) and O( P) channels to
2
00
2
00
Fig. 4. When NO photodissociates at the photon
2
produce NO(X P, v ¼ 0) and NO(X P, v ꢀ 6),
respectively. Our result obtained from the one-
photon dissociation experiment is consistent with
their result derived from the two-photon dissoci-
ation.
0
0
3
energy lower than Elimit(J NO), O( P) should be
2
produced as a counter product of NO(X P1=2,
0
0
00
v ¼ 0, J ). The clear observation of the
NO(X P1=2, v ¼ 0, J ) production threshold at
the corresponding Elimit(J NO) means that the
2
00
00
0
0
Fig. 5 shows the lifetimes obtained from the
lineshape analysis in this work and the available
literature data as a function of the excitation en-
ergy [2,12]. The literature data are available only
2
00
00
counter product of NO(X P1=2, v ¼ 0, J ) is
1
O( D). This experimental result provides a rigor-
ous evidence showing that NO
~
in the D(1,0,0)
2
2
1
À1
level predissociates into NO( P) and O( D).
Uselman and Lee [6] measured the neopentyl
alcohol production yield through the insertion
for the energy region below 42 000 cm . The
agreement between the three data sets seems quite
good. One set of data are determined by a reso-
nance Raman method, and probably provide the
short-time limits of the lifetimes [12]. The line-
1
reaction of neopentane with O( D) as a function of
1
NO
is produced after the excitation through the D B
2
photolysis energy, and suggested that O( D)
2
~
2
–
bands located above the threshold energy of
width of the (0,0,0) band was measured to be
À1
~
2
X A
1
0.127 cm
from the absorption spectrum ob-
À1
4
0 996 cm . Our spectroscopic result accords well
tained at room temperature [2], which is almost
the same value as ours. It should be mentioned
0 0
Fig. 5. Lifetimes of various (m ; m ; m ) levels obtained from this
1 2
0
Fig. 4. PHOFEX spectra of the (1,0,0) band obtained by
0
3
0
monitoring various J values of a photodissociation product
work and the available literature as a function of excitation
energy: Ã, this work by LIF/hole-burning/PHOFEX; , ab-
2
00
NO(X P1=2, v ¼ 0). The broken lines indicate the simulated
ꢁ
Lorentzian profile of Fig. 3. The bold arrows indicate the dis-
0
sorption [2]; }, resonance Raman [12]. The bold arrow indi-
0
00
sociation limit Elimit(J NO) to form a specific rotational level of
cates the dissociation limit Elimit(J NO ¼ 0) to form a specific
2
00
00
1
2
00
00
1
NO(X P1=2, v ¼ 0, J ) and O( D).
rotational level of NO(X P1=2, v ¼ 0; J ¼ 0) and O( D).

606
K. Tsuji et al. / Chemical Physics Letters 374 (2003) 601–607
0
0
that the rotational level (N ¼ 9, K ¼ 3) analyzed
below the second dissociation limit. This obser-
vation implies that the predissociation rate into the
in absorption spectrum by Hallin and Merer is
0
0
3
different from those (N ¼ 1, 3, 5 of the K ¼ 0
manifold) in our experiment. This implies that the
lifetimes are independent of the rotational quan-
tum number; in other words, the predissociation
proceeds through homogeneous interaction. The
O( P) channel is high enough to compete with
1
the O( D) channel in the energy region above the
second dissociation limit. The competition of the
predissociation channel results in almost the same
1
3
values of quantum yields of O( D) and O( P)
above the second dissociation limit.
~
2
homogeneous interaction between the D B and
2
~
2
A B
2
states is one of the probable candidate to
explain the predissociation mechanism of the
present interest.
It is found that all the vibronic bands studied by
the PHOFEX method have extremely broad line-
widths, and the corresponding lifetimes are shorter
than that of the (0,0,0) band. More importantly,
the steep decrease in lifetime occurs beyond the
In summary, the LIF, hole-burning, and
PHOFEX spectra were measured for jet-cooled
NO
lifetime for NO
vestigated. The lifetimes for the vibrational levels
2
and the vibrational level dependence of the
~
2
2 2
in the D B state has been in-
~
2
2
in the D B state are estimated from the line pro-
files of the LIF, hole-burning, and PHOFEX
spectra, which increase steeply as the excitation
energy increases even below the second dissocia-
(
0,0,0) band. The lifetime reaches about 100 fs at
À1
2
1
the (0,0,2) level, which is located 152 cm below
the second dissociation limit, and no marked en-
ergy dependence has been noticed in the energy
region above the (0,0,2) level. It is clearly demon-
strated that the second dissociation limit is not
important to determine the lifetime. This trend of
lifetimes indicates that the predissociation rate to
tion limit to form NO( P1=2) and O( D). The
appearance thresholds were observed in the
PHOFEX spectra obtained by monitoring various
00
J -levels of nascent NO, which accord with the
2
calculated energies required for forming NO( P1=2
,
00
00
1
v ¼ 0, J ) and O( D). This observation clearly
1
indicates the occurrence of the O( D) dissociation
2
2
3
~
form NO( P) and O( P) increases sharply as the
vibrational energy increases in the D B state.
2
2
pathway via the D B state.
~
2
Roughly speaking, the predissociation rate of
the (0,0,2) levels is 260 times faster than that of
the (0,0,0) level. If the predissociation occurs via
Acknowledgements
~
2
2
the A B state as mentioned before, the linewidth
This work was partly supported by a Grant-
in-Aid for Scientific Research on Priority Areas
C of the peaks is expressed in the FermiÕs golden
2
rule as C ¼ 2pW q, where W is the interaction
(B) ÔPhotochemistry of molecular complexesÕ
(No. 13127202) from the Ministry of Education,
~
2
~
2
2 2
term between the D B and A B states, and q is
~
the density of states in the A B
2
2
state. The density
state is quite high because energy gap
Culture, Sports, Science and Technology of
Japan.
~
of the A B
2
2
~
between the (0,0,0) levels in the D B
2
~
and A B
2
2
2
À1
states is as large as 30 394 cm . At least we can
safely say that the density of states of the coupled
states does not change drastically over the excita-
References
À1
tion energy region of 40 000–41 000 cm , where
[
[
[
[
[
1] R.K. Ritchie, A.D. Walsh, P.A. Warsop, Proc. R. Soc.
London. Ser. A 266 (1962) 257.
we recognize a sharp rise in the predissociation
rate.
This trend of the predissociation rate seems
2] K.-E.J. Hallin, A.J. Merer, Can. J. Phys. 54 (1976)
1157.
3] K. Aoki, K. Hoshina, K. Shibuya, J. Chem. Phys. 105
1
consistent with that of the O( D) quantum yield
obtained by Uselman and Lee [6]. The quantum
(
1996) 2228.
4] K. Shibuya, C. Terauchi, M. Sugawara, K. Aoki, K. Tsuji,
S. Tsuchiya, J. Mol. Struct. 413–414 (1997) 501.
5] R. Jost, J. Nyg ꢀa rd, A. Pasinski, A. Delon, J. Chem. Phys.
105 (1996) 1287.
1
yield of O( D) is almost constant around 0.5 above
the second dissociation limit. The predissociation
13
À1
rate reaches about 10
s at (0,0,2), which locates

K. Tsuji et al. / Chemical Physics Letters 374 (2003) 601–607
607
[
[
[
[
6] W.M. Uselman, E.K.C. Lee, J. Chem. Phys. 65 (1976)
948.
7] T.G. Slanger, W.K. Bischel, M.J. Dyer, J. Chem. Phys. 79
1983) 2231.
8] L. Bigio, R.S. Tapper, E.R. Grand, J. Phys. Chem. 88
1984) 1271.
9] L. Bigio, E.R. Grant, J. Chem. Phys. 87 (1987) 5589.
[15] H.-S. Im, E.R. Bernstein, J. Phys. Chem. A 106 (2002)
7565.
1
[16] R.E. Smalley, L. Wharton, D.H. Levy, J. Chem. Phys. 63
(1975) 4977.
(
[17] G. Inoue, Y. Nakata, Y. Usui, H. Akimoto, M. Okuda,
J. Chem. Phys. 70 (1979) 3689.
(
[18] W.N. Sisk, C.E. Miller, H.S. Johnston, J. Phys. Chem. 97
(1993) 9916.
[
[
10] J. McFarlane, J.C. Polanyi, J.G. Shapter, J. Photochem.
Photobiol. A: Chem. 58 (1991) 139.
[19] M. Kawasaki, K. Kasatani, H. Sato, H. Shinohara, N.
Nishi, Chem. Phys. 78 (1983) 65.
11] N. Shafer, K. Tonokura, Y. Matsumi, S. Tasaki, M.
Kawasaki, J. Chem. Phys. 95 (1991) 6218.
[20] A.M. Bass, A.E. Ledford Jr., A.H. Laufer, J. Res. Nat.
Bur. Stand. 80A (1976) 143.
[
[
12] R. Fan, L.D. Ziegler, J. Raman Spectrosc. 25 (1994) 497.
13] M. Ahmed, D.S. Peterka, A.S. Bracker, O.S. Vasyutinskii,
A.G. Suits, J. Chem. Phys. 110 (1999) 4115.
[21] J.B. Coon, F.A. Cesani, F.P. Huberman, J. Chem. Phys. 52
(1970) 1647.
[
14] R.C. Richter, V.I. Khamaganov, A.J. Hynes, Chem. Phys.
Lett. 319 (2000) 341.
[22] L. Bigio, E.R. Grant, J. Phys. Chem. 89 (1985) 5855.
[23] L. Bigio, E.R. Grant, J. Chem. Phys. 87 (1987) 360.