First spectroscopic observation of gas-phase HOONO

Article abstract of DOI:10.1021/jp013598l

The three-body association reaction of hydroxyl radical with NO₂, OH + NO₂ + M → HONO₂ + M is one of the most important processes in the chemistry of the Earth's lower atmosphere. The first direct spectroscopic observation of HOONO formed in the reaction of OH with NO₂ was reported. Rich vibrational structure, consistent with the existence of several HOONO conformers, was observed. A tentative vibrational assignment of the observed bands was suggested, although the complete assignment is not possible without further spectroscopic information. The ratio HOONO/HONO₂ formed in the reaction of OH with NO₂ was ～ 5% at 253 K and 20 torr.

Full text of DOI:10.1021/jp013598l

© Copyright 2002 by the American Chemical Society
VOLUME 106, NUMBER 6, FEBRUARY 14, 2002
LETTERS
First Spectroscopic Observation of Gas-Phase HOONO
Sergey A. Nizkorodov*,† and Paul O. Wennberg‡
California Institute of Technology, Pasadena, California 91125
ReceiVed: September 21, 2001; In Final Form: NoVember 28, 2001
A vibrational overtone photodissociation spectrum of HOONO formed in the reaction of OH with NO2 is
reported. Rich vibrational structure, consistent with the existence of several HOONO conformers, is observed.
A tentative vibrational assignment of the observed bands is proposed although complete assignment is not
possible without further spectroscopic information. The ratio HOONO/HONO2 formed in the reaction of OH
with NO2 is estimated to be about 5% at 253 K and 20 Torr.
Introduction
the product branching ratio in reaction 1 and the unimolecular
decomposition rate of the implicated HOONO intermediate can
have a drastic effect on the apparent rate of OH disappearance
in kinetic experiments, such as polyexponential and/or isotope
specific OH removal.5
The three-body association reaction of hydroxyl radical with
nitrogen dioxide
OH + NO2 + M f HONO2 + M
(1a)
Golden and Smith4 analyzed the available kinetic data for
reaction 1 using a RRKM approach and concluded that the k1b/
k1a branching ratio may be as high as 15% under typical
atmospheric conditions, and substantially larger at higher total
pressures. On the contrary, Matheu and Green6 concluded that
reaction 1b is insignificant under all relevant atmospheric
conditions. Probably the strongest experimental evidence for
reaction 1b was presented in a recent report by Donahue et al.,5
who studied the kinetics of reaction 1 using isotopically labeled
OH. Indirect experimental evidence of HOONO formation was
also obtained from the observation of a double-exponential
decay of OH in excess NO2.7 A direct spectroscopic technique
for observation of HOONO could clearly provide a much more
rigorous experimental constraint for the mechanism of reaction
1. Much to the frustration of experimentalists, however, two
major attempts to detect HOONO in reaction 1 and, in fact, in
several other reactions thought to produce HOONO in the gas
phase were unsuccessful.8,9
is one of the most important processes in the chemistry of the
Earth’s lower atmosphere.1 By sequestering HOx and NOx into
the more chemically inert and photochemically stable nitric acid,
this reaction terminates several important catalytic cycles
involving odd-hydrogen and odd-nitrogen radicals, and directly
affects the stratospheric and tropospheric ozone budgets.
Detailed knowledge of both the mechanism and kinetics of
reaction 1a is, therefore, of critical significance for understanding
atmospheric photochemistry.
Although the kinetics of reaction 1 has been a subject of
numerous studies under a variety of pressure and temperature
conditions,2,3 there still is a surprising amount of disagreement
between studies performed at high and low pressure. It has been
speculated4 that part of this disagreement may arise from a
possible reversible side branch to reaction 1,
OH + NO2 + M T HOONO + M
(1b)
Direct observation of HOONO vibrational bands in solid
Argon matrixes10-12 and strong evidence for the transient
existence of HOONO in aqueous solutions13-17 proves beyond
suspicion that this molecule is not a figment of the theoretical
chemists’ imagination. Indeed, HOONO is expected to be bound
by as much as 18.8 kcal/mol18 and have at least three stable,
which results in a weakly bound adduct of OH and NO2. Indeed,
† Camille and Henry Dreyfus Postdoctoral Scholar in Environmental
Chemistry; nizkorod@caltech.edu.
‡ Division of Geological & Planetary Sciences and Division of Engineer-
ing and Applied Science; wennberg@gps.caltech.edu.
10.1021/jp013598l CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/18/2002

856 J. Phys. Chem. A, Vol. 106, No. 6, 2002
Letters
Figure 2. Photodissociation spectrum of HOONO in the first OH
stretching overtone region. The strongest bands at 6365 and 6935 cm-1
are tentatively assigned to the 2ν1 vibrations of the cis-cis and {cis-
perp + trans-perp} conformers of HOONO, respectively. The corre-
sponding anticipated positions of HOONO overtones estimated from
the matrix isolation spectra are indicated by vertical bars. The spectrum
of PNA, which consists of a single strong 2ν1 band at 6900 cm-1 and
weaker combination bands at higher frequencies, is also shown for
comparison.
Figure 1. Lowest energy conformers of HOONO. The structures and
relative energies of the cis-cis, cis-perp, and trans-perp conformers (in
kcal/mol) are taken from ab initio calculations of ref 19. In this notation,
the first and second labels refer to the ONOO and NOOH torsional
angles, respectively.
nearly isoenergetic conformers differing mostly in the magni-
tudes of their equilibrium HOON and OONO torsional angles18-20
(Figure 1). The failure to observe HOONO in the previous
experiments must therefore have chemical or dynamical origins.
In this Letter, we report the first direct spectroscopic
observation of HOONO produced in reaction 1 and an experi-
mental estimate of the branching ratio k1b/k1a. This measurement
takes advantage of the vibrational predissociation spectroscopy
approach famous for its high degree of sensitivity and selectivity
in spectroscopic studies of weakly bound van der Waals and
ionic complexes.21,22 The main idea is to use vibrational overtone
excitation in assisting the decomposition of HOONO back into
the original components:
OH is produced in a microwave discharge of H2 in a large
excess of either helium or argon. The flow coming out of the
microwave cavity is mixed with a flow of 0.7% NO2 in N2 and
directed in either a FTIR spectrometer or photolysis/LIF
chamber maintained at 253 K. The flow conditions are chosen
to ensure that all H-atoms are converted into OH
H + NO2 f OH + NO
(3)
and all OH radicals are tied up with NO2 in reaction 1 before
they reach the OH LIF detection region. Under typical condi-
tions, optimized for the largest IR-induced signal-to-background
ratio, the initial mixing ratios are 0.1-0.2% NO2 and 0.04-
0.1% H2 at a total helium-nitrogen ballast pressure of 15-25
Torr. A bubbler with an aqueous solution of PNA is used in
place of the microwave setup to calibrate the intensity of the
near-IR vibrational photodissociation signal against the known
photodissociation cross-sections of PNA.23
HOONO + hν f HOONO(2ν1) f OH + NO2 (2)
wherein the resulting OH photofragments are sensitively
detected with laser-induced fluorescence (LIF) against es-
sentially no background other than the unreacted OH.
Results and Discussion
Experimental Section
Figure 2 displays a photodissociation spectrum observed by
detecting OH fluorescence while scanning the frequency of the
near-IR excitation laser. The spectrum consists of at least seven
partly overlapping bands distributed between 6000 and 7200
cm-1, i.e., in the frequency range characteristic of the first OH
stretching overtones. The occurrence of photodissociation at
these excitation frequencies implies that the parent molecule is
weakly bound (D0 < ≈7000 cm-1). The signal requires all
components of the experiment including the microwave dis-
charge, pump and probe lasers, and H2 and NO2 flows. Adding
NO to the mixture to convert possible HO2 photofragments into
OH
The experimental setup is similar to the one described in our
recent publication on the near-infrared photolysis of peroxynitric
acid (PNA ≡ HO2NO2)23 and only the most pertinent informa-
tion is provided here. A tunable pulsed OPO laser (2 cm-1
resolution; 1 mJ/pulse; 0.3 cm beam diameter) supplies radiation
in the frequency range of the first (2ν1) and second (3ν1) OH
2
stretching overtones. OH is detected by LIF in the A2Σ r Π
band excited by a copropagating tunable dye laser near 282 nm
(,1 µJ/pulse; 0.2 cm beam diameter) in a single photon
counting regime. For time-resolved pump-probe experiments,
the pulse repetition rate of both lasers is set at 100 Hz and the
delay between them is scanned. In the integration mode, which
is occasionally used for acquiring spectra, a 6400 Hz comb of
probe laser pulses is generated, shifted from the pump laser
comb by a known variable delay. In both cases, the gated
counting window of the photon counter is locked onto the probe
laser pulse, with the gate width (2-4 µs) comfortably accom-
modating several OH fluorescence lifetimes (≈500 ns24). To
minimize the saturation of the counting electronics the count
rate is kept below 0.05-0.1 counts/gate.
HO2 + NO f OH + NO2
(4)
does not affect the signal, indicating that the molecule(s)
responsible for the observed spectrum must directly produce
OH upon absorption of IR radiation. We explicitly verified that
no known stable impurities or secondary reaction products, such
as HNO3, HONO, H2O, etc. produce an IR-induced OH signal
under comparable conditions (e.g., via a two-photon vibra-

Letters
J. Phys. Chem. A, Vol. 106, No. 6, 2002 857
TABLE 1: Estimated Gas-Phase Band Frequencies (cm-1
of the OH Stretching Vibrations of HOONO Conformersa
)
modes of HOONO. For example, the bands at 6505 and 6630
cm-1 could be combinations of cis-cis 2ν1 with ν9 and 2ν9,
where ν9 is the NOOH torsional mode. Similarly, the bands at
7045 and 7155 cm-1 could be the corresponding combinations
for the cis-perp/trans-perp conformers. The NOOH torsional
mode is predicted by ab initio calculations to carry a large
oscillator strength and, depending on the conformer, have
ν1(exp) ν1(exp)
ν1(est) 2ν1(est)
molecule
gasb Ar matrixb matrix/gas gas
gas
cis-cis HOONO
cis-perp HOONO
trans-perp HOONO
HNO3
HO2
t-HONO
3285
3298 6430
3554 6940
3574 6980
3545
3521
3413
3558
3412
3550
3436
3591
3426
0.9918
0.9933
0.9908
0.9959
harmonic frequencies of 200-400 cm-1 18-20,27
Provided that
.
our assignment is correct, the much lower experimental values
for ν9 frequencies (140 cm-1 for the cis-cis conformer and 110
cm-1 for the -perp ones) would indicate a large degree of
structural “floppiness” in HOONO. However, in the absence
of more reliable information on HOONO spectroscopy, such
assignments should be regarded as highly tentative.
c-HONO
a The fractional matrix shifts for the cis-cis (0.9959) and trans-perp
(0.9920) HOONO fundamentals are taken as the shift for c-HONO and
the average of HNO3, HO2, and t-HONO shifts, respectively. The
fundamental frequency of the cis-perp conformer is estimated from that
of the trans-perp conformer and the ab initio frequency shift (20 cm-1
)
Predicting relative intensities is especially problematic. Ac-
cording to the ab initio calculations of McGrath and Rowland,19
the relative energies of the three stable conformers of HOONO
are 0 (cis-cis), 0.7 (cis-perp), and 2.8 (trans-perp) kcal/mol. The
corresponding predicted relative intensities for the OH stretching
fundamentals are quite similar (32, 40, and 44 km/mol,
respectively). Assuming that the scaling between the overtone
and fundamental intensities is the same for all conformers, and
using ideal gas thermodynamic properties of HOONO calculated
in ref 19, one predicts that the relative 2ν1 intensities at 253 K
should be 1.0, 0.45, and 0.008 for the cis-cis, cis-perp, and trans-
perp HOONO, respectively. The observed intensity pattern (1.0
for cis-cis and 5.0 for the combination of cis-perp with trans-
perp) appears to be disturbingly different from the estimated
one. The band intensity for the trans-perp isomer is especially
at odds with the predictions considering its quite noticeable
contribution to the width of the 6935 cm-1 band and also the
fact that this conformer was the first to be discovered in the
matrix.12
between them.19 Anticipated overtone band centers are obtained from
the estimated OH fundamental frequencies using an anharmonicity
constant of 83 cm-1 (average for OH, HNO3, and HO2NO2). b Refer-
ences 12, 26, and 28.
TABLE 2: Observed Band Positions (cm-1) in the Overtone
Photodissociation Spectrum of HOONO Shown in Figure 2a
band center
((15 cm-1
)
rel intensity
tentative assignment
6180
6365
6505
6630
6730
6935
7045
7155
0.1
1
2ν1 (cis-cis)
0.5
0.2
0.9
5b
2ν1 + ν9 (cis-cis)
2ν1 + 2ν9 (cis-cis)
2ν1 (cis-perp & trans-perp)
2ν1 + ν9 (cis-perp & trans-perp)
2ν1 + 2ν9 (cis-perp & trans-perp)
0.6b
0.2
a The specified band assignment (ν1 ) OH stretch; ν9 ) NOOH
torsion) must be regarded as highly tentative. The assignment for the
strong band at 6730 cm-1 is presently unclear. b Overlap with each
other.
The photodissociation spectrum shown in Figure 2 is,
however, a convolution of the absorption cross section with the
presently unknown dissociation quantum yield. The dissociation
energy of the cis-cis conformer of HOONO is calculated to be
tionally mediated dissociation by the IR pump and UV probe
lasers25). PNA does photofragment into HO2 and NO2 in this
excitation frequency range, but its spectrum, shown in Figure
2 for comparison, is very different from the spectrum observed
here. On the basis of the available spectroscopic and theoretical
evidence,10,12,18,19 we argue below that HOONO conformers are
the carriers of the bands shown in Figure 2.
6580 cm-1 18
, i.e., higher than its presumed 2ν1 band origin (6365
cm-1). The intensity of the cis-cis 2ν1 band may therefore be
suppressed relative to the intensities of higher frequency
vibrations. The assumption of equal scaling between 2ν1 and
ν1 intensities is also highly questionable, especially when
comparing hydrogen-bonded with free OH stretches. In addition,
HOONO conformers are known to be separated from each other
by sizable barriers,5,8 possibly preventing complete equilibration
among them on the time scale of our experiment. Finally, the
exponential sensitivity of the predicted intensities on the relative
energies of HOONO conformers makes such predictions less
trustworthy. A reexamination of the relative stabilities of
HOONO conformers may help to resolve these issues.
In the absence of any data on gas-phase spectroscopy of
HOONO,26 the complete assignment of the observed overtone
spectrum is a challenging task. The only existing infrared spectra
of HOONO were observed in matrix isolation studies after UV
irradiation of HNO3 in solid argon.10-12 Bands at 3285 and 3545
cm-1 were assigned to the OH stretching fundamental vibrations
of cis-cis and trans-perp conformers (Figure 1) of HOONO,
respectively. The third conformer of HOONO, cis-perp, was
not observed in the matrix although it is calculated to have an
intermediate stability compared to the other two.19 Using the
available matrix data, combined with typical values for OH
matrix shifts and vibrational anharmonicities, the first OH
stretching overtones can be estimated as 6430, 6940, and 6980
cm-1 for cis-cis, cis-perp, and trans-perp conformers of HOONO,
respectively (see Table 1). The estimated overtone positions
closely match the positions of the strongest features in the
experimental spectrum (Figure 2 and Table 2) suggesting that
it arises from photofragmentation of HOONO. Based on our
simple analysis, we tentatively assign the band at 6365 cm-1
to the 2ν1 transition of the cis-cis conformer, and the anoma-
lously wide 6935 cm-1 band to the overlapping 2ν1 transitions
of cis-perp and trans-perp isomers. The remaining bands are
likely to be combinations of 2ν1 with low-frequency vibrational
From the point of view of atmospheric chemistry, the product
branching ratio k1b/k1a is the most interesting quantity. Although
we cannot provide a rigorous quantitative measurement of this
ratio, we are able to estimate it from relative measurements of
PNA vs HOONO photolysis rates under well-characterized
conditions. To do this, a known amount of PNA (as measured
by FTIR spectroscopy) is admitted in the photolysis chamber
and both the photolysis spectrum of PNA and the OH time
profile are recorded. The PNA flow is then replaced by the
H-NO2 flow and the corresponding data are taken for HOONO
under essentially unchanged conditions (the LIF probe laser
power is increased by a calibrated amount to account for the
weaker signal from HOONO). The concentration of H atoms
making it from the microwave discharge region to the mixing

858 J. Phys. Chem. A, Vol. 106, No. 6, 2002
Letters
on experimental conditions. In addition, the unimolecular life-
time of HOONO has to be quantified before the [HOONO]/
[HNO3] ratio obtained here for a fixed time delay (0.3 s)
between mixing and observation can be related to the actual
branching ratio k1b/k1a. We are presently in a process of
modifying our apparatus to make it suitable for flow tube
kinetics experiments, both to put the preliminary measurements
described here on a more quantitative basis and to investigate
the chemistry of HOONO in more detail.
In summary, this Letter presents the first observation of a
gas-phase spectrum of HOONO, a structural isomer of nitric
acid that has long eluded experimentalists. Although a complete
vibrational analysis of the spectrum is not feasible at the present
time, both the number of the observed bands and their positions
are consistent with the available spectroscopic information on
this molecule. We have shown that the fraction of HOONO
produced in the association reaction of OH with NO2 is relatively
small, of the order of a few percent. It remains to be seen
whether this product channel becomes more significant at lower
temperatures and higher pressures. From an experimental
kinetics perspective, the successful observation of HOONO by
photodissociation spectroscopy under ambient temperature
conditions opens the door to many interesting experiments on
reactions involving weakly bound adducts such as, for example,
ROONO and RO2-H2O (R ) organic radical).
Figure 3. Kinetic traces of OH produced in the infrared photolysis of
HOONO (open circles) and PNA (filled circles) as a function of the
pump-probe delay. The signals have not been corrected for the
difference in probe laser powers. The primary product of PNA
photolysis is HO2, which is converted into OH via reaction with NO.
Kinetic data shown here, in conjunction with photolysis spectra of
HOONO and PNA taken under well-defined conditions, are used to
estimate the product branching ratio in reaction 1.
point is quantified using NO2 titration, allowing for calculation
of the final HNO3 density. The [HNO3] and [NO2] concentra-
tions are additionally verified using FTIR spectroscopy under
similar flow conditions and found to agree with the results of
titration measurement to within 20%.
Figure 3 shows sample kinetic traces of OH concentrations
following the near-IR photolysis of PNA and HOONO. Pho-
tolysis of PNA produces HO2, which is converted into the
detected OH via reaction 4 with NO. The rise time of the
observed trace is consistent with the rate of reaction 4, and the
decay is almost entirely due to the reaction of OH with PNA.
Photolysis of HOONO generates OH on the time scale of the
pump laser pulse, consistent with OH being the primary
photoproduct. The OH decay in this case is mostly due to
reaction 1. With this information at hand, the relative initial
amounts of HO2 and OH produced in the photolysis of PNA
and HOONO is calculated in a straightforward manner. As-
suming that the integrated absorption transition strength is the
same for PNA and HOONO overtones and correcting for the
nonunity dissociation quantum yield of PNA (0.16 at 253 K),23
we can estimate the density of HOONO in the photolysis
chamber.
We estimate, using the method described above, that
[HOONO]/[HONO2] ) 0.05 ( 0.03 at 253 K in 20 Torr
helium-nitrogen buffer approximately 0.3 s after the mixing
of OH and NO2. The uncertainties do not include systematic
errors due to the assumptions described above, some of which
are not fully substantiated. For example, HOONO dissociation
quantum yield of less than 1 would tend to increase the ratio.
On the other hand, our inclusion of the HOONO combination
bands in the integrated signal decreases the ratio somewhat. The
integrated absorption cross sections for HOONO and PNA are
likely to be different, but probably not by more than 50%. But
even combining all these uncertainties together, we do not expect
the actual ratio to be off by more than a factor of 2. Thus, the
results presented here imply an approximate lower limit of 5 (
3% for the branching ratio k1b/k1a.
Acknowledgment. This work was funded in part by NASA’s
Atmospheric Effects of Aviation Program (NAG5-11157) and
the National Science Foundation’s Atmospheric Chemistry
Program (ATM-0094670). The OPO laser system was developed
with partial support from the National Science Foundation’s
Major Research Instrumentation Program (ATM-9724500;
Geoffrey A. Blake, PI). S.A.N. thanks the Camille and Henry
Dreyfus Foundation for support.
References and Notes
(1) Finlayson-Pitts, B. J.; Pitts, J. N. Chemistry of the Upper and Lower
Atmosphere: Theory, Experiments, and Applications; 1999.
(2) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling: EValuation Number 12; JPL: Pasadena, CA, 1997.
(3) 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.
(4) Golden, D. M.; Smith, G. P. J. Phys. Chem. A 2000, 104, 3991.
(5) Donahue, N. M.; Mohrschladt, R.; Dransfield, T. J.; Anderson, J.
G.; Dubey, M. K. J. Phys. Chem. A 2001, 105, 1515.
(6) Matheu, D. M.; Green, W. H., Jr. Int. J. Chem. Kinet. 2000, 32,
245.
(7) Hippler, H.; Nasterlack, S.; Striebel, F.; Golden, D. M. The Reaction
of OH + NO2 + M: Rate constants and branching ratios for isomer
formation. Proceeding of the 5th International Conference on Chemical
Kinetics; NIST: Gaithersburg, MA, 2001.
(8) Dransfield, T. J.; Donahue, N. M.; Anderson, J. G. J. Phys. Chem.
A 2001, 105, 1507.
(9) Burkholder, J. B.; Hammer, P. D.; Howard, C. J. J. Phys. Chem.
1987, 91, 2136.
(10) Lo, W.-J.; Lee, Y. P. J. Chem. Phys. 1994, 101, 5494.
(11) Koch, T. G.; Sodeau, J. R. J. Phys. Chem. 1995, 99, 10824.
(12) Cheng, B. M.; Lee, J. W.; Lee, Y. P. J. Phys. Chem. 1991, 95,
2814.
Our result is marginally consistent with the upper limit for
HOONO yield obtained by Burkholder et al. (5%)9 but overlaps
well with the upper limit from the work of Dransfield et al.
(10%).8 The isotope-specific kinetics study of reaction 1 by
Donahue et al.5 and calculations by Golden and Smith4 appear
to be on a higher side of the ratio reported here. However, one
has to be careful when comparing different experiments with
each other, as the yield of HOONO is likely to depend strongly
(13) Coddington, J. W.; Hurst, J. K.; Lymar, S. V. J. Am. Chem. Soc.
1999, 121, 2438.
(14) Richeson, C. E.; Mulder, P.; Bowry, V. W.; Ingold, K. U. J. Am.
Chem. Soc. 1998, 120, 7211.
(15) Loegager, T.; Sehested, K. J. Phys. Chem. 1993, 97, 6664.
(16) Merenyi, G.; Lind, J. Chem. Res. Toxicol. 1998, 11, 243.
(17) Kissner, R.; Nauser, T.; Bugnon, P.; Lye, P. G.; Koppenol, W. H.
Chem. Res. Toxicol. 1997, 10, 1285.
(18) Li, Y.; Francisco, J. S. J. Chem. Phys. 2000, 113, 7976.

Letters
J. Phys. Chem. A, Vol. 106, No. 6, 2002 859
(19) McGrath, M. P.; Rowland, F. S. J. Phys. Chem. 1994, 98, 1061.
(20) Sumathi, R.; Peyerimhoff, S. D. J. Chem. Phys. 1997, 107, 1872.
(21) Okumura, M.; Yeh, L. I.; Lee, Y. T. Springer Ser. Opt. Sci. 1985,
(24) Chidsey, I. L.; Crosley, D. R. J. Quant. Spectrosc. Radiat. Transf.
1980, 23, 187.
(25) Hausler, D.; Andresen, P.; Schinke, R. J. Chem. Phys. 1987, 87,
3949.
49, 122.
(22) Bieske, E. J.; Dopfer, O. Chem. ReV. 2000, 100, 3963.
(23) Roehl, C. M.; Nizkorodov, S. A.; Zhang, H.; Blake, G. A.;
Wennberg, P. O. J. Phys. Chem., in press.
(26) Melen, F.; Herman, M. J. Phys. Chem. Ref. Data 1992, 21, 831.
(27) Chakraborty, D.; Park, J.; Lin, M. C. Chem. Phys. 1998, 231, 39.
(28) Jacox, M. E. J. Chem. Phys. Ref. Data Monograph No. 3 1994.