Radiative lifetimes of the FeO orange system

Article abstract of DOI:10.1016/S0009-2614(00)00294-3

The ground-state FeO molecules are generated from photolysis of Fe(CO)₅ in a Fe(CO)₅/M(O₂ or N₂O)/Rg(He or Ar) mixture using an unfocused weak UV laser beam. The formation of ground-state FeO molecules is identified by a laser-induced fluorescence (LIF) method. The LIF signal from FeO molecules is stronger in O₂ than in N₂O at the same partial pressures. The radiative lifetimes for seven bands in the FeO orange system are measured. They are substantially different depending on the excited band ranging from 260±30 ns to 590±50 ns.

Full text of DOI:10.1016/S0009-2614(00)00294-3

14 April 2000
Ž
.
Chemical Physics Letters 320 2000 658–664
www.elsevier.nlrlocatercplett
Radiative lifetimes of the FeO orange system
)
H.S. Son, K. Lee, S.K. Shin, J.K. Ku
Department of Chemistry, Pohang UniÕersity of Science and Technology, San 31 Hyoja-Dong Pohang, 790-784, Kyung-buk, South Korea
Received 24 January 2000; in final form 17 February 2000
Abstract
Ž
.
Ž
.
Ž
.
Ž
.
The ground-state FeO molecules are generated from photolysis of Fe CO in a Fe CO rM O2 or N2O rRg He or Ar
5
5
mixture using an unfocused weak UV laser beam. The formation of ground-state FeO molecules is identified by a
Ž
.
laser-induced fluorescence LIF method. The LIF signal from FeO molecules is stronger in O2 than in N2O at the same
partial pressures. The radiative lifetimes for seven bands in the FeO orange system are measured. They are substantially
different depending on the excited band ranging from 260"30 ns to 590"50 ns. q 2000 Elsevier Science B.V. All rights
reserved.
1. Introduction
In this Letter, we like to report a simple method
of generating the ground-state FeO molecules and
the results of radiative lifetime measurements for the
FeO orange system. We have previously reported
that a state-selected excited state of Fe atoms can be
There have been substantial interests in the FeO
molecule because of its complicated electronic states
with six d electrons and its potential astrophysical
w x
importance of this molecule 1 . Various experimen-
Ž
.
generated by multiphoton dissociation MPD of
tal techniques have been employed to generate FeO
molecules in the gas phase; they involve reactions of
O2 with Fe atoms generated from various sources,
Ž
.
Fe CO vapor at Fe atomic transition frequencies
5
w
x
using an unfocused weak UV laser pulse 10,11 . As
a continuing research, we attempted to investigate
w x
w
such as arc discharge of an iron electrode 2 , a
the reaction of a state-selected excited-state Fe atoms,
x
high-temperature oven containing metallic iron 3,4 ,
w x
3
Ž
.
Fe z D3 for an example, with O2 and N2O by
hallow cathode discharge 5 , burning ferrocene,
Ž
.
Ž
.
Ž
photolyzing Fe CO molecules in a Fe CO rM O2
or N2O rRg He or Ar mixture at 319.16 nm. Strong
5
5
Ž
.
Fe C5H5 2 , in a 2450-MHz microwave discharge
.
Ž
.
w
x
w
x
6,7 , and laser ablation method 8,9 . Most of the
3
Ž
.
atomic emissions originated from Fe z D3 state were
easily observed, but no chemiluminescence from
FeOU was detected. We pursued two-color photoly-
sis-probe experiments to look for the formation of
the ground-state FeO molecules under our experi-
mental conditions. When a visible probe laser was
tuned to excite the FeO molecules following the
previous works have been focused on the spectro-
scopic characterization of the ground-state FeO
molecule and the FeO orange system. However,
radiative lifetimes of the FeO orange system have
w x
been rarely studied 4 .
)
Ž
.
Ž
.
irradiation of a Fe CO rM O2 or N2O rHe mix-
Corresponding author. Tel.: q82-562-279-2107; fax: q82-
5
ture with an unfocused weak UV laser pulse, rela-
562-279-3399; e-mail: jkku@vision.postech.ac.kr
0009-2614r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 00 00294-3

(
)
H.S. Son et al.rChemical Physics Letters 320 2000 658–664
659
Ž
.
tively strong laser-induced fluorescence LIF signals
from the FeO orange system were observed. This
finding of an easy way to generate ground-state FeO
molecules in the gas phase enabled us to study the
radiative lifetimes and quenching rate constants for
seven bands of the FeO orange system.
graphic grating and a Hammamatsu R928 photo-
Ž
.
multiplier PM tube. The signal from the PM tube
Ž
was digitized with a transient digitizer Tektronix
7912HB and transferred to a laboratory computer
.
for signal averaging and storage.
3. Results
2. Experimental method
3.1. Excitation and emission spectra
Fig. 1a shows a low resolution excitation spec-
An experimental setup used for this work is basi-
w
x
cally the same as that reported previously 10 . The
reaction cell was made of a 1-liter pyrex bulb and
two pairs of 2.5 cm pyrex O-ring joints were at-
tached to the bulb for the laser beam path and for the
connection to a gas handling vacuum rack. A pre-
trum of FeO molecules in the 555–615 nm region,
U
Ž
.
which corresponds to the FeO orange system
5
Ž
.
w
x
FeO X D4 transitions 6,7 . The photolysis wave-
length was 319.16 nm, which corresponds to the
Ž
.
mixture of 0.75% Fe CO in He or Ar was slowly
5
flowed through the cell. When the flow was stabi-
lized at a certain pressure, the other reactant gas
Ž
.
pure O2 or N2O was introduced directly to the cell
through a needle valve installed on a separated line.
A typical flow rate of gases in the reaction cell was
;0.2 mmol miny1. The total pressure as well as the
partial pressures of the gases in the cell was con-
trolled by adjusting the openings of inlet needle
valves and an exit teflon valve. When the pressure in
the cell was stabilized again, unfocused UV laser
pulses in the 298–320 nm region were directed to
the cell. The pulse energy of the photolysis laser
Ž
.
Quantel YG681-TDL50 with NBP was 0.5;1.0
mJrpulse and the beam diameter was ;7 mm.
5
Ž
.
The ground-state FeO X D4 molecules were
probed by LIF signals from the FeO orange system.
Ž
The probe laser Quantel YG581-TDL60 with NBP
.
and DGO was scanned in the 555–615 nm region
and fluorescence from the FeO orange system was
monitored at longer wavelengths shifted about 880
cmy1 from the excitation frequencies. The monitor-
U
Ž
ing wavelengths correspond to the FeO orange sys-
5
Y
.
Ž . w x
tem
FeO X D4, Õ s1 transitions 6,7 . A typi-
cal delay time between the photolysis and probe was
set at ;20 ms to avoid interference from the Fe
atomic fluorescence caused by the photolysis laser
pulse. The pulse energy and linewidth of the probe
laser was ;0.3 mJrpulse and ;0.05 cmy1, re-
spectively. Fluorescence from the excited FeO
molecules was monitored at the right angle with
respect to the laser beam direction by using a 50 cm
Ž .
Fig. 1. a A low-resolution excitation spectrum of FeO generated
Ž
.
from photolysis of a 0.75% Fe CO 5, 25% O2 and ;74% He
Ž . Ž .
Ž
.
558.3,
mixture. Total pressure was 0.4 Torr.
Ž . Ž .
A
B
579.0,
Ž .
G 611.0 nm band.
C
Ž .
D
582.0,
Ž .
b
586.8,
Typical time resolved fluorescence spectra of FeO observed
Ž . Ž .
E
591.9,
F
600.5, and
U
Ž
.
from 591.1 nm band excitation: i 30–130, ii 130–230, and iii
230–330 ns period. The composition of the gas mixture and total
X
Y
Ž .
pressure are the same as those in a . The intensities of the v
1
Ž
.
monochromator Spex 1870C equipped with a holo-
peak are normalized and the normalization factors are shown.

(
)
660
H.S. Son et al.rChemical Physics Letters 320 2000 658–664
z3D3 a5D4 transition of the Fe atom. While scan-
ning the probe laser in the 555–615 nm region, the
fluorescence from the excited states was monitored
.
N2O rHe mixture are shown in Fig. 2a. The excita-
tion wavelength was 591.1 nm, and the fluorescence
was observed at 623.6 nm. Both time profiles clearly
show single exponential decay. Since the partial
Ž . Ž .
Ž .
Ž .
Ž .
at A 587.2, B 610.2, C 613.6, D 618.5, E
Ž .
Ž .
Ž
. Ž
.
623.6, F 634.1, and G 645.8 nm, respectively.
pressure of Fe CO ;2 mTorr is much lower than
5
The total pressure was 0.4 Torr and the gas mixture
those of other gases, the decay rate of the 623.6 nm
Ž
.
was composed of 0.75% Fe CO , 25% O2 and
emission can be written as
5
X
about 74% He or Ar. The excitation spectrum clearly
shows characteristic features of FeO orange system
indicating the formation of the ground-state FeO
molecules in the reaction cell. When only one laser
was employed, no fluorescence from the FeO orange
system was detected, ensuring that FeO molecules in
the orange system were not formed directly by side
reactions. Small changes in the photolysis wave-
length from the Fe atomic transition frequency in-
duced no significant changes in the shape and rela-
tive intensities of the excitation spectra, suggesting
that contributions from the excited state Fe atoms
generated by the photolysis laser seemed not impor-
tant. However, the LIF intensities from the FeO
orange system became much stronger when the pho-
tolysis wavelength was moved to shorter wavelength.
Fig. 1b shows time resolved fluorescence spectra
obtained from excitation of the 591.1 nm band head
w x
w
x
1rts1rt0 qkq M qkq He .
1
Ž .
M represents O2 or N2O, t0 is the radiative lifetime,
kq and kXq are the quenching rate constants. Both kq
and kXq were determined by varying the pressure of
one component while keeping other component con-
stant. In order to determine kq for O2 or N2O, the
Ž
.
pressure of a 0.75% Fe CO in He mixture was kept
5
constant at 0.3 Torr. To determine kXq for He or Ar,
the pressure of O2 was kept constant at 0.1 Torr and
the pressure of the rare gases containing 0.75%
5
X
Y
Ž
.
Ž .
of the D4
Õ
X5D4 Õ s0 transition. The pres-
sure and compositions of the gas mixture were iden-
tical to those of Fig. 1a. The fluorescence signal was
collected after 30 ns from the probe laser pulse to
avoid the effect of the scattered laser light. The time
intervals were arbitrarily chosen to examine the vari-
ation of relative emission intensities of the two peaks
in different time domains. The relative emission
intensities of the two peaks in Fig. 1b remain con-
stant in different time periods, suggesting that they
come from the common upper state. The energy
spacing between the excitation frequency and the
first strong emission peak is 882"10 cmy1, and
that between the two successive emission peaks is
866"9 cmy1. These energy spacings match well
with those of the ground-state vibrational levels re-
w
x
ported by Merer and coworkers 6,7 , where the two
X
Y
X
Y
Ž
.
Ž
.
emission peaks are assigned for 0 , 1 and 0 , 2
bands, respectively.
Fig. 2.
following 558.3 nm excitation in O2 I and N2O ` . The total
Ž .
Typical time profile of FeOU observed at 587.3 nm
Ž .
a
Ž
.
Ž .
3.2. RadiatiÕe lifetimes of the FeO orange system
pressure was 0.4 Torr. b Pressure dependence of decay frequen-
Typical time profiles observed from the FeO or-
Ž
.
Ž .
cies for the emission observed at 587.3 nm; He e , Ar ^ , O2
Ž
.
Ž
Ž
.
Ž .
ange system at 0.4 Torr of a Fe CO rM O2 or
B , and N2O ` .
5

(
)
H.S. Son et al.rChemical Physics Letters 320 2000 658–664
661
Table 1
Ž
.
5
Fe CO was varied in the 0.2–1.1 Torr range. Since
Radiative lifetimes and quenching rate constants for FeO orange
system
Ž
.
the concentration of Fe CO was very low, contri-
bution of Fe CO to the quenching rate constants
5
Ž
.
5
a
b
Ž
.
FeO band to ns
Quenching kq
gas
for He or Ar was neglegible. The pressure depen-
dence of the decay rates obtained in this way for the
558.3 nm band is plotted in Fig. 2b, and those for
other bands are shown in Fig. 3. Since the zero-pres-
sure intercepts in Fig. 2b and Fig. 3 contain the
quenching rate from the fixed component, the radia-
tive lifetimes are obtained by substracting the calcu-
lated quenching rate of the fixed component from the
zero-pressure intercept. The radiative lifetimes and
quenching rate constants by O2 , N2O, He and Ar for
seven bands of the FeO orange system are shown in
Table 1. The radiative lifetimes and apparent quench-
611.0 nm
590"50
He
Ar
O2
N2O
He
Ar
O2
He
Ar
1.4"0.1
2.8"0.2
2.7"0.2
3.6"0.3
1.8"0.1
3.1"0.3
3.0"0.3
0.90"0.05
3.1"0.2
3.3"0.3
4.2"0.2
0.95"0.05
2.9"0.2
2.8"0.2
3.4"0.2
0.84"0.03
3.1"0.2
3.4"0.2
4.0"0.2
0.79"0.03
600.5 nm
591.1 nm
360"40
280"30
O2
N2O
He
Ar
586.8 nm
582.0 nm
579.0 nm
558.3 nm
260"30
560"50
460"50
O2
N2O
He
Ar
O2
N2O
He
Ar
c
c
Ž
.
Ž
.
450"100
2.5"0.2 0.034
2.4"0.2
3.7"0.2
0.94"0.04
2.8"0.2
O2
N2O
He
Ar
260 " 30
c
Ž
.
;500
O2
2.6"0.2
N2O
3.6"0.2
a The error limits are 2s from the average values obtained
from the intercepts in Fig. 2b and Fig. 3.
b Units: 10y10 cm3 moleculey1 sy1
w x
Ref. 4 .
.
c
ing rate constants measured for the 558.3 and 586.8
nm bands are virtually the same, because the upper
level for these two bands is known to be the same
w x
one 7 . In general, the magnitude of the apparent
quenching rate constants by Ar and O2 is very
similar, but that by N2O is about 25% larger than
that by O2.
3.3. Power dependence and optical pumping
Fig. 3. Pressure dependence of decay frequencies for the emis-
Ž .
Ž .
Ž .
Ž .
Ž .
sions from
a
579.0,
b
582.0,
c
586.8,
d
591.1, e 600.5,
Ž .
and f 611.0 nm band excitation. The emissions were observed at
Ž . Ž . Ž . Ž . Ž . Ž .
To look for the origin of the ground-state FeO
molecules, we have investigated photolysis laser
power dependence of LIF intensities from the FeO
molecules as well as the Fe atoms under the same
experimental conditions as the excitation spectra ex-
a 610.2, b 613.6, c 618.5, d 623.6, e 634.1, and f 645.8
Ž
.
Ž
.
Ž
.
Ž .
nm, respectively; He e , Ar ^ , O2 B , and N2O ` . The
pressure in the abscissa represents either the added pressures of
O2 or N2O to the 0.3 Torr of 0.75% Fe CO in He, or those of
0.75% Fe CO in He or Ar to the 0.1 Torr of O2.
Ž
.
5
Ž
.
5

(
)
662
H.S. Son et al.rChemical Physics Letters 320 2000 658–664
cept the photolysis laser wavelength. Since the FeO
LIF signal was much stronger at shorter photolysis
laser wavelength, the sample gas mixture was pho-
tolyzed at 298.0 nm. For FeO molecules, the probe
laser wavelength was at 558.3 nm and the LIF
intensities were monitored at 587 nm. For Fe atoms,
We also employed an optical pumping method to
examine changes in FeO LIF intensity with increase
in concentration of the metastable Fe atomic species.
5
Ž
When the gas mixture was photolyzed at the Fe y F5
5
.
Ž
.
a D4 atomic transition frequency 296.7 nm , the
5
Ž
.
LIF intensity from Fe a F5 state was three times
stronger than that observed from photolyzing the gas
mixture using a nearby off-resonance wavelength
the probe laser was at y3D3 a3F4 381.58 nm and
Ž
.
y5D4 a5F5 382.04 nm transition frequencies for
the a3F4 and a5F5 states, respectively, and the LIF
Ž
.
Ž
.
296.6 nm because of the optical pumping of the
intensities were monitored at y3D3 a3F3 390.2
a5D4 state Fe atoms to the a5F5 state through the
Ž
nm and y D4 a5D3 305.9 nm transitions for the
a3F4 and a5F5 states, respectively. Because of the
fast disappearance of a3F4 and a5F5 states of Fe
atoms, the atomic LIF signal was monitored after 10
ms from the photolysis pulse. The LIF intensities
from the FeO orange bands were first-order as shown
short lived y F5 state ts8.7 ns . However, the LIF
intensities from the FeO molecules remained con-
stant within the experimental fluctuations when the
photolysis laser wavelength was scanned in the
296.6–297 nm region. Optical pumping for other
levels in the a5F and a3F terms also attempted, and
similar results were obtained. Thus, these metastable
states of Fe atoms do not contribute appreciably, if
any, to the formation of FeO molecules observed in
this work.
5
5
.
Ž
.
Ž
.
Ž
in Fig. 4a, whereas those from atomic species Fe
3
5
.
a F4 and a F5 showed second-order dependence as
plotted in Fig. 4b and c. Thus, the ground-state FeO
molecules generated in our experimental conditions
seemed to be originated from the reaction of
Ž
.
photofragments of Fe CO and oxidant molecules.
5
4. Discussion
The spectroscopic properties of the ground state
and orange system of FeO have been established by
w
x
Merer and coworkers 6,7,12 . They have shown that
5
Ž
there are two excited electronic states D D and
X5
.
w x
D D in this spectral region 12 , and reassigned
some of the vibrational assignments reported by
w x
West and Broida 4 . The LIF spectrum shown in
Fig. 1a, which is consistent with those spectra from
FeO reported by Merer and coworkers, manifest
itself as the direct evidence of the formation of
ground-state FeO molecules in the reaction cell.
Also, the first-order dependence of the FeO LIF
intensities on the photolysis laser power strongly
suggests that FeO molecules are formed from the
Ž
.
Ž
.
reaction of O2 or N2O with Fe CO
xs1–4
x
produced by the photolysis laser pulse. It is found
that the radiative lifetimes of the FeO orange system
are substantially different from band to band. The
radiative lifetime for the 579.0 nm band measured in
this work agrees well with the one reported by West
3
Ž .
Ž .
Ž
.
Ž .
Fig. 4. Photolysis laser power dependence of a FeO, b Fe a F4 ,
w x
and Broida 4 , while that for the 558.3 nm band is
5
Ž .
1.1"0.1,
Ž
.
and c Fe a F5 LIF intensity. The magnitude of the slope is a
b 2.0"0.1 and c 2.0"0.1, respectively. The gas
mixture was composed of 0.3 Torr of 0.75% Fe CO in He and
quite different from the previous one.
Ž . Ž .
The chemical reactivity of Fe atoms with various
small molecules near room temperature has been
Ž
.
5
0.1 Torr of O2.

(
)
H.S. Son et al.rChemical Physics Letters 320 2000 658–664
663
w
x
w
x
studied by Mitchell and Hackett 13 . They observed
states of Fe atoms by various small molecules 19 .
5
Ž
.
that the removal rate of the ground-state Fe a D4
They reported the depletion rate constants for
5
Fe a F5 atoms by O2 and N2O to be 8.3=10y11
and 3.7=10y11 cm3 moleculey1 sy1, respectively.
We also measured the depletion rate constants for
Ž
.
atoms did not increase significantly over the diffu-
sion rate in the presence of O2 or N2O, thereby
concluding that the ground-state Fe atoms were not
reactive toward O2 or N2O. We checked for the
depletion kinetics of the ground-state Fe atoms by
using a photolysis-probe method and reached the
5
Ž
.
the Fe a F5 atoms by O2 and N2O, and obtained
8.6"0.3=10y11 and 3.0"0.2=10y11 cm3 mole-
culey1 sy1, respectively, which are in good agree-
ment with Mitchell and Hackett. They assigned the
depletion rate constants for the metastable Fe atoms
to the collisional relaxation processes based on ki-
netic modeling studies, since they did not succeed to
observe LIF signals from FeO at 578.9 nm. Our
observation of the formation of the ground-state FeO
in this work, however, does not change Mitchell and
Hackett’s assignment of large depletion rate con-
stants for the collisional relaxation processes, be-
cause we observe almost no enhancement of FeO
LIF intensity in spite of increasing the concentration
of metastable Fe atoms by optical pumping.
w
x
same conclusion as Mitchell and Hackett 13 . In
fact, the reaction of the ground-state Fe atoms with
Ž
O2 Eq.
w
is endothermic by 84–92 kJ moly1
Ž ..
2
x
Ž .
3,13–15 , so that the reaction in Eq. 2 is very
unlikely to occur under our experimental conditions.
Fe a5D , g qO
g
2 Ž .
FeO g qO.
Ž .
2
Ž .
Ž
.
4
On the other hand, the reaction of the ground-state
Fe atoms with N2O, Eq. 3 , is highly exothermic by
Ž .
x
;247 kJ moly1 13,14,16 .
Fe a5D , g qN O g
w
FeO g qN g .
Ž . 2 Ž .
3
Ž .
Ž .
Ž
.
4
2
Ryther and Weitz have studied reaction kinetics
xs1–4 photo-fragments
in the gas phase 20 . They have shown that the
triplet state of Fe CO reacts much faster with O2
having triplet ground state than H2 or CO having
singlet ground state. Since the ground states of
Ž .
Ž
.
Ž
.
But it has been reported that the reaction in Eq. 3
does not proceed in an appreciable rate at low tem-
perature 13,16 . The activation energy for the reac-
of unsaturated Fe CO
w
x x
w
x
Ž
.
4
Ž .
tion in Eq. 3 has been reported to be 44.4"1.0
and 45.1"2.3 kJ moly1 by Campbell and Metzger
w
x
w x
Ž
.
Ž
.
w x
16 and Zaslonko and Smirnov 17 , respectively.
Fe CO
xs1–4 photofragments are triplet 21 ,
x
Thus, the ground-state Fe atoms are not considered
to be the source of ground-state FeO molecules
observed in this work. Since the LIF signal from the
FeO orange system is absent unless the photolysis
the much stronger FeO LIF intensity observed in an
O2 mixture in comparision with that in a N2O
mixture seems consistent with the spin selection rule
proposed by Weitz and coworkers 20 . It seems that
reactions of Fe CO
w
.
x
Ž
.
Ž
.
Ž
laser pulse is preceded, the direct reaction of Fe CO
with O2 or N2O in the gas phase is also ruled out for
xs1–4 photofragments,
5
x
Ž
.
most likely those highly unsaturated Fe CO or
2
the source of the ground-state FeO molecules. As the
FeCO, with O2 and N2O are responsible for the
production of FeO in this work.
5
Fe a FJ state lies 6928–8154 cmy1 ;83–97 kJ
Ž
.
Ž
moly1 above the ground state 18 , it is nearly
thermoneutral or slightly exothermic to form
.
w x
Acknowledgements
ground-state FeO molecules in reactions with O2.
3
The Fe a FJ atoms lying 11 976–12 969 cmy1
Ž
.
Ž
;
This work is financially supported in part by the
143–155 kJ moly1 above the ground state are also
energetically viable for the formation of ground-state
FeO molecules. Nevertheless, both metastable states
of the Fe atoms are also ruled out as the source of
the ground-state FeO, because the FeO LIF intensi-
ties remain unchanged in spite of increasing the
concentrations of these states by optical pumping.
Mitchell and Hackett have measured the bimolecular
removal rate constants for a 5D3, a5F5,4 and a3F4,3
.
Ž
Korea Science and Engineering Foundation 96-
.
0501-02-01-3 , and in part by the Ministry of Educa-
Ž
.
tion BSRI-97-3438 .
References
w x
1
M.D. Allen, L.M. Ziurys, J.M. Brown, Chem. Phys. Lett.
Ž
.
257 1996 130.
w x
2
Ž .
R.F. Barrow, M. Senior, Nature 223 1969 1359.

(
)
664
w x
H.S. Son et al.rChemical Physics Letters 320 2000 658–664
Ž
.
w
x
3
w x
4
A. Fontijn, S.C. Kurzius, Chem. Phys. Lett. 13 1972 507.
12 M. Barnes, M.M. Fraser, P.G. Hajigeorgiou, A.J. Merer, S.D.
Ž
.
Ž .
Rosner, J. Mol. Spectrosc. 170 1995 449.
J.B. West, H.P. Broida, J. Chem. Phys. 62 1975 2566.
w x
Ž
.
w
w
x
Ž
.
5
S.M. Harris, R.F. Barrow, J. Mol. Spectrosc. 84 1980 334.
A.S.-C. Cheung, N. Lee, A.M. Lyyra, A.J. Merer, A.W.
13 S.A. Mitchell, P.A. Hackett, J. Chem. Phys. 93 1990 7822.
w x
x
14 JANEF Thermochemical Tables, 3rd edn., Dow Chemical,
6
Ž
.
Ž .
Midland, MI, J. Phys. Chem. Ref. Data 14, Suppl. 1 1985 .
Taylor, J. Mol. Spectrosc. 95 1982 213.
w x
7
w
x
A.S.-C. Cheung, A.M. Lyyra, A.J. Merer, A.W. Taylor, J.
15 G. Bulducci, G. de Maria, M. Guido, V. Piacente, J. Chem.
Ž
.
Ž .
Phys. 55 1971 2596.
Mol. Spectrosc. 102 1983 224.
w x
8
w
x
Ž
.
M. Barnes, M.M. Fraser, P.G. Hajigeorgiou, A.J. Merer, S.D.
16 M.L. Campbell, J.R. Metzger, Chem. Phys. Lett. 253 1996
Ž
.
Rosner, J. Mol. Spectrosc. 170 1995 449.
158.
w x
9
w
x Ž .
17 I.S. Zaslonko, V.N. Smirnov, Kinet. Catal. 49 1998 485.
G.V. Chertihin, W. Saffel, J.T. Yustein, L. Andrew, M.
Neurock, A. Ricca, C.W. Bauschlicher Jr., J. Phys. Chem.
w
x
18 C.E. Moore, Atomic Energy Levels, Vol. 2, NSRDS-NBS
Ž
.
100 1996 5261.
Circular No. 35, US GPO, Washington, DC, 1971.
w
w
x
Ž
.
w
w
w
x
Ž
.
10 K. Lee, J.S. Goo, J.K. Ku, Chem. Phys. Lett. 244 1995
19 S.A. Mitchell, P.A. Hackett, J. Chem. Phys. 93 1990 7813.
x
Ž
.
213.
20 R.J. Ryther, E. Weitz, J. Phys. Chem. 95 1991 9841.
x
x Ž .
21 I.M. Waller, J.W. Hepburn, J. Chem. Phys. 88 1988 6658.
11 J.S. Goo, K. Lee, S.C. Bae, J.K. Ku, J. Chem. Phys. 105
Ž
.
1996 7485.