Rate constants of the O(1D) reactions with N2, O 2, N2O, and H2O at 295 K

Article abstract of DOI:10.1016/j.cplett.2005.05.062

Using a technique of laser flash photolysis coupled with vacuum ultraviolet laser-induced fluorescence spectroscopy, the rate coefficients of O(¹D) reactions with N₂, O₂, N₂O, and H₂O at 295 ± 2 K have been determined to be kN2=(3.29±0.27)×10-11, kO2=(4.06±0.24)×10-11, kN2O=(1.35±0.08)×10-10 and kH2O=(2.07±0.18)×10-10cm3molecule-1s-1. The quoted uncertainties include estimated errors and are the 95% confidence level. The kN2 and kN2O values obtained are larger than the current NASA/JPL recommendations by 26% and 16%, respectively, although they are still within the error limits associated with the recommendations.

Full text of DOI:10.1016/j.cplett.2005.05.062

Chemical Physics Letters 410 (2005) 196–200
www.elsevier.com/locate/cplett
Rate constants of the O(1D) reactions with
N2, O2, N2O, and H2O at 295 K
*
Kenshi Takahashi , Yukari Takeuchi, Yutaka Matsumi
Solar-Terrestrial Environment Laboratory, Graduate School of Science, Nagoya University, 3-13 Honohara, Toyokawa, Aichi 442-8507, Japan
Received 14 April 2005; in final form 23 May 2005
Available online 14 June 2005
Abstract
Using a technique of laser flash photolysis coupled with vacuum ultraviolet laser-induced fluorescence spectroscopy, the rate
coefficients of O(1D) reactions with N2, O2, N2O, and H2O at 295 2 K have been determined to be kN ¼ ð3.29 ꢀ 0.27Þꢁ
2
10ꢂ11, kO ¼ ð4.06 ꢀ 0.24Þ ꢁ 10ꢂ11, kN O ¼ ð1.35 ꢀ 0.08Þ ꢁ 10ꢂ10 and kH O ¼ ð2.07 ꢀ 0.18Þ ꢁ 10ꢂ10 cm3 moleculeꢂ1
s
ꢂ1. The quoted
2
2
2
uncertainties include estimated errors and are the 95% confidence level. The kN and kN O values obtained are larger than the current
NASA/JPL recommendations by 26% and 16%, respectively, although they are still within the error limits associated with the
2
2
recommendations.
ꢀ 2005 Elsevier B.V. All rights reserved.
1. Introduction
Therefore, there are now large kinetic databases con-
cerning chemical reactions involving O(1D) at ambient
temperatures [3,4]. A variety of experimental techniques
have been utilized to investigate the kinetics of the atmo-
spheric O(1D) reactions [5–18]. The O(1D) concentration
can be monitored by emission spectroscopy (ES) at
630 nm associated with O(1D–3P) transition [8,9]. It is
also possible to measure the O(1D) concentration by res-
onance absorption (RA) via the 3s1D–2p1D transition at
115.2 nm [7]. When the branching ratio for O(3P) pro-
duction is significant in O(1D) reactions, the total reac-
tion rate constants of O(1D) can be determined from
the temporal evolution rate of the O(3P) products. The
O(3P) atoms can be monitored by RA spectroscopy
and resonance fluorescence (RF) spectroscopy via the
3s3S0–2p3Pj transition at 130 nm [12,15–17]. A micro-
wave-powered oxygen atom lamp has been utilized to
obtain VUV radiation at 130 nm in those experiments.
Recently, we have demonstrated that laser-induced fluo-
rescence (LIF) spectroscopy via the 3s1D–2p1D transi-
tion at 115.2 nm is a powerful technique to study the
chemical processes in the atmosphere [19–28]. For in-
stance, photodissociation processes of atmospheric O3
In the terrestrial atmosphere, O(1D) atom is mainly
produced from the photolysis of O3 in the ultraviolet re-
gion [1–4]. Although the abundance is very low, a small
fraction of O(1D) reactions create highly reactive species
from highly unreactive species; this minor pathway for
O(1D) loss is often the major pathway for the generation
of the reactive species. Specifically, OH radical in the
stratosphere and troposphere and NO (and eventually
all nitrogen oxides) in the stratosphere are produced
mostly from the reactions of O(1D) with inert H2O
and N2O.
1
Oð DÞ þ H2O ! 2OH
ð1Þ
ð2Þ
1
Oð DÞ þ N2O ! 2NO
The reactive species OH and NO are immensely impor-
tant in the atmosphere [1–4].
*
Corresponding author. Fax: +81 533 89 5593.
E-mail address: kent@stelab.nagoya-cu.ac.jp (K. Takahashi).
0009-2614/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2005.05.062

K. Takahashi et al. / Chemical Physics Letters 410 (2005) 196–200
197
in the UV region have extensively been studied by the
LIF detection of O(1D) atoms [21–24,26–28]. The ki-
netic measurements for O(1D) reactions have also been
performed through the LIF detection of O(1D) [18,25].
Based on critical reviews of those experimental stud-
ies, both NASA/JPL [5] and IUPAC [6] databases in-
clude the recommendation values of the kinetic data
for use in atmospheric studies. Very recently, some
experimental studies which have revisited the O(1D)
reactions with atmospheric molecules have suggested
that there are discrepancies in the rate constants be-
tween the currently recommended values and newly
measured values [15–18]. For instance, three different
groups at National Oceanic and Atmospheric Adminis-
tration (NOAA), Georgia Institute of Technology, and
University of Leeds, have independently conducted the
kinetic studies to measure the rate constant of
(8.95 · 10ꢂ20 cm2 moleculeꢂ1) [5], the photolysis laser
fluence (1.4 · 1014 photons cmꢂ2), and the O(1D) quan-
tum yield from N2O photolysis at 193 nm (ꢃ1) [5,6].
Tunable VUV radiation around 115.22 nm was gen-
erated by frequency tripling of UV laser light in Ar
phase-matched Xe [29]. The 345.6 nm laser light
(ꢃ7 mJ/pulse) was focused into the cell containing the
Xe/Ar mixture with a fused silica lens (f = 170). The typ-
ical partial pressures of Xe and Ar were 30 and 60 Torr,
respectively. Photolysis and probe laser beams crossed
at right angles in the reaction chamber. The distance be-
tween the probe beam entrance window and the crossing
point of the two lasers was 70 mm. The distance between
the crossing point of the two lasers and the window
where the fluorescence exits the cell was 60 mm. The
O(1D) LIF was detected by a solar-blind photomulti-
plier tube (EMR, 541J-08-17), in which the direction
of the LIF observation was orthogonal to the propaga-
tion direction of both VUV probe and photolysis laser
beams. Output from the PMT was pre-amplified and
fed into a gated integrator (Stanford Research, SR-
250). Temporal decay profiles of O(1D) as a function
of the delay time between the photolysis and probe laser
pulses were thus measured in the experiments. The delay
time was controlled by a digital pulse generator (Stan-
ford Research, DG535).
O(1D) + N2 reaction (kN ) at 295 K [15]. They deter-
2
mined the kN value to be 3.1 · 10ꢂ11 cm3 mole-
2
culeꢂ1 sꢂ1, which is larger than the latest NASA/JPL
recommendation by 19%. The difference strongly affects
modeling studies of the production rates of OH radical
in the atmosphere, because the OH radical is formed
mainly through reaction (1) and the steady-state concen-
tration of atmospheric O(1D) is predominantly depen-
dent on kN and kO . Therefore, it is obvious that
2
2
further kinetic studies are required to investigate the dis-
crepancies revealed by the recent studies.
The reaction chamber was evacuated continuously
using a rotary pump through a liquid nitrogen trap.
Pressure in the chamber was monitored by a capacitance
manometer (MKS, Baratron622). Total pressure in the
chamber was maintained at 5 Torr with an excess of
He which was added to thermalize the velocity distribu-
tion of O(1D) generated from the photolysis of N2O
within 500 ns [19]. Physical quenching of O(1D) by He
was negligible under our experimental conditions,
because the room-temperature rate constant for
O(1D) + He was reported to be 7 · 10ꢂ16 cm3 mole-
culesꢂ1 sꢂ1 [30]. The reactant concentrations in the
chamber were estimated using the mass flow rates and
pressures. All the mass flow meters were calibrated using
a primary mass flow calibrator (STEC, SF-1). The
uncertainties in the mass flow rate and pressure mea-
surements were estimated to be 2% for N2O, N2, O2,
and He. For H2O experiments, a stable flow of water va-
por was produced by bubbling a flow of He through a
liquid sample of distilled H2O maintained at 290 K.
The absolute concentration of H2O was determined
using a thermo-hygrometer (Shinyei Co., TRH-CA).
The uncertainty in the H2O concentration measure-
ments was estimated to be 4%. Purities of N2, O2,
N2O and He were 99.99995% (Japan Fine Products
Co.), 99.9% (Nihon Sanso Co.), 99.9% (Air Liquid Ja-
pan Co.), and 99.9999% (Air Liquid Japan Co.), respec-
tively. They were used in the experiments without
further purification. All experiments have been per-
formed at 295 2 K.
In this Letter, we report the experimental determina-
tions of the rate constants for O(1D) reactions with N2,
O2, N2O and H2O at 295 2 K. Time-resolved experi-
ments to measure the decay profiles of O(1D) in the pres-
ence of reactants reveal the rate constants, in which the
relative concentration of O(1D) has been monitored by
LIF spectroscopy at 115.2 nm. Results are compared
with the recent kinetic measurements by other groups
and the NASA/JPL and IUPAC recommendations.
2. Experimental
The experimental apparatus used in the present study
is almost the same as in our previous studies on the pho-
tochemical processes involving O(1D) [19–28]. Pulsed la-
ser-flash photolysis combined with time-resolved LIF
was used in all experiments to investigate the reaction
kinetics of O(1D). Reactant mixtures of N2O and colli-
sion partners (N2, O2, N2O, and H2O) and a large excess
of Helium are slowly introduced into a reaction chamber
through calibrated mass flow meters (STEC, SEC-400).
Photolysis of N2O at 193 nm was used to generate the
O(1D) atoms, in which the 193 nm laser light was ob-
tained by an Excimer laser (Lambda Physik, Optex).
The initial concentration of O(1D), [O(1D)]t = 0, was
about 2 · 1011 atoms cmꢂ3, which was estimated from
the reported absorption cross section of N2O

198
K. Takahashi et al. / Chemical Physics Letters 410 (2005) 196–200
3. Results and discussion
6
5
4
3
2
1
0
Fig. 1 shows a trace of the observed temporal profile
of O(1D) LIF signal at 115.2 nm following the 193 nm
pulsed laser irradiation of a mixture of 25 mTorr of
N2O and 200 mTorr of O2 in 5 Torr of He diluent.
The time-resolved VUV–LIF signal of O(1D) atoms pro-
duced by the photolysis of N2O exhibits an initial jump
followed by a slower decay. The subsequent decay
follows pseudo-first-order kinetics and provides infor-
mation on the kinetics of O(1D) + O2 reaction at
295 K. We have estimated an uncertainty of 2% in
the pseudo-first-order decay rates for O(1D) decay attrib-
utable to the non-linear least squares fitting method and
the choice of the range of data used here.
0.0
0.4
0.8
1.2
1.6
[Reactant] (1016molecules cm-3)
Fig. 2. Plot of pseudo-first-order loss of O(1D) atoms versus the
concentration of N2 (rectangular), O2 (triangle), H2O (rhombus), and
N2O (circle). Slopes are the results of linear weighted fit analysis of the
experimental data and indicate the bimolecular rate constants for those
O(1D) reactions at 295 2 K.
Fig. 2 shows a plot of the observed pseudo-first-order
decay of O(1D) in the presence of N2, O2, N2O, and H2O
reactants. Straight lines drawn in Fig. 2 are the results of
linear least squares fit analysis. The slopes of the straight
lines give the bimolecular rate constants of (3.29
0.27) · 10ꢂ11, (4.06 0.24) · 10ꢂ11, (1.35 0.08) ·10ꢂ10
and (2.07 0.18) · 10ꢂ10 cm3 moleculeꢂ1 sꢂ1 for O(1D)
reactions with N2, O2, N2O, and H2O, respectively.
Quoted uncertainty in the bimolecular rate constants is
two standard deviations, which includes the uncertainty
in the slope of a plot of the pseudo-first-order decay
rates versus the reactant concentration (Fig. 2) and the
estimated systematic uncertainties in the measurements.
The systematic uncertainties are due to the uncertainties
in the concentration of the excess reagents and the statis-
tical uncertainties in the values of the pseudo-first-order
rate constants derived from the exponential fitting.
Table 1 lists the room-temperature bimolecular rate
constants of the O(1D) reactions, in which both the re-
sults from the present study and previous studies are in-
cluded for comparison. For O(1D) + N2 reaction, the
presently determined kN value is in agreement with
2
the recent reports by Ravishankara et al. [15], Strekow-
ski et al. [16], Dunlea and Ravishankara [17], Blitz et al.
[18], within the uncertainties. Although the five recent
results are still within the error limits of the current
NASA/JPL recommendation, they are larger than the
recommendation systematically. For O(1D) + O2 reac-
tion, the presently determined kO value is in good agree-
2
ment with the available literature values within the
quoted uncertainties, as listed in Table 1. For O(1D) +
H2O reaction, our value is in good agreement with the
values reported by Amimoto et al. [10], Gericke and
Comes [13], Wine and Ravishankara [12], and Dunlea
and Ravishankara [17], within the quoted uncertainties.
Those five data agree with the current NASA/JPL and
IUPAC recommendations within the quoted uncertain-
ties, but are slightly smaller than the recommendations
systematically.
OH radicals in the lower and middle atmosphere are
predominantly produced by the reaction of H2O with
O(1D) which is generated by O3 photolysis. Due to its
pivotal role in the atmosphere, accurate estimation of
the atmospheric OH production rate is an important
issue [2–4]. The atmospheric OH production rate,
P(OH), is expressed as follows:
Fig. 1. A typical trace of the LIF signal of O(1D) as a function of the
delay time between the N2O photolysis and O(1D) probe laser pulses.
The inset shows a plot of ln(LIF intensity) versus the delay time. The
O(1D) atoms are detected using a technique of vacuum UV laser-
induced fluorescence (VUV–LIF) at 115.22 nm corresponding to the
3s1D0–2p1D transition. Reactant pressures are 25 mTorr, 200 mTorr,
and 5 Torr for N2O, O2 and He, respectively. Experiments were
performed at 295 2 K.
2 kH O½H2Oꢄ
2
PðOHÞ ¼
kH O½H2Oꢄ þ kN ½N2ꢄ þ kO ½O2ꢄ
2
2
2
ꢁ JðO1DÞ ꢁ ½O3ꢄ
ð3Þ
In Eq. (3), kX are the rate constants for O(1D) reactions
with X, and J(O1D) is the rate of O(1D) production

K. Takahashi et al. / Chemical Physics Letters 410 (2005) 196–200
199
Table 1
Summary of the room-temperature rate constants of O(1D) reactions with N2, O2, N2O, and H2O
b
Reactant
N2
Rate constanta
Detected species
Experimental techniquec
Reference
3.29 0.27
3.24 0.3
3.00 0.24
3.06 0.25
3.1 0.2
O(1D)
O(3P)
O(3P)
O(1D)
O(1D),
O(3P)
O(1D)
O(3P)
O(3P)
O2ðb1Rþg Þ
LIF
RF
RF
LIF
LIF
RF
ES
This work
Strekowski et al. [16]
Dunlea and Ravishankara [17]
Blitz et al. [18]
Ravishankara et al. [15]
2.8 0.6
2.4 0.1
2.52 0.25
2.6 0.3
2.6 1.0
2.6 0.4
Streit et al. [8]
RA
RF
ES
Amimoto et al. [10]
Wine and Ravishankara [12]
Shi and Barker [14]
Sander et al. [5]
Atkinson et al. [6]
O2
4.06 0.24
4.2 0.4
3.89 0.25
3.8 0.4
4.1 0.5
4.2 0.2
4.0 0.8
4.0 0.5
O(1D)
O(3P)
O(3P)
O(1D)
O(1D)
O(3P)
LIF
RF
RF
LIF
ES
This work
Strekowski et al. [16]
Dunlea and Ravishankara [17]
Blitz et al. [18]
Streit et al. [8]
RA
Amimoto et al. [10]
Sander et al. [5]
Atkinson et al. [6]
N2O
1.35 0.08
1.27 0.08
1.07 0.10
1.1 0.2
O(1D)
O(3P)
O(1D)
O(1D)
O(3P)
O(3P)
LIF
RF
LIF
ES
This work
Dunlea and Ravishankara [17]
Blitz et al. [18]
Davidson et al. [9]
Amimoto et al. [10]
Wine and Ravishankara [12]
Sander et al. [5]
1.2 0.1
RA
RF
1.17 0.12
1.16 0.35
1.16 0.2
Atkinson et al. [6]
H2O
2.07 0.18
1.96 0.18
2.3 0.5
O(1D)
O(3P)
O(1D)
O(3P)
O(3P)
OH
LIF
RF
ES
This work
Dunlea and Ravishankara [17]
Streit et al. [8]
1.95 0.3
2.6 0.5
RA
RA
RA
RF
Amimoto et al. [10]
Lee and Slanger [11]
Gericke and Comes [13]
Wine and Ravishankara [12]
Sander et al. [5]
2.02 0.41
1.95 0.2
2.2 0.4
O(3P)
2.2 0.5
Atkinson et al. [6]
a
In units of 10ꢂ11 cm3 moleculeꢂ1 sꢂ1 for N2 and O2 reactions, and 10ꢂ10 cm3 moleculeꢂ1 sꢂ1 for N2O and H2O reactions. Quoted uncertainties
associated with this work are two standard deviations including the uncertainty in the slope of a plot of the pseudo-first-order decay rates versus the
reactant concentration and the estimated systematic uncertainties in the measurements.
b
Species detected for kinetic measurements.
LIF, laser-induced fluorescence spectroscopy; RF, resonance fluorescence spectroscopy using an oxygen atom lamp; RA, resonance absorption
spectroscopy; ES, emission spectroscopy.
c
from O3 photolysis. The P(OH) value is very sensitive to
kX values. Using the rate constants determined in the
present study, we have estimated the fractional change
in the calculated OH production rate relative to that cal-
culated from the current recommendations. Although
the uncertainty in the OH production rate calculated
from the recommendations is relatively large (ꢅ35% at
the 67% confidence level) due to propagation of the
uncertainties in the rate constants, the OH production
rate calculated from our new rate constants is roughly
15% smaller than that from the recommendations. Re-
cent studies have suggested that the calculated OH pro-
duction rates become smaller throughout the lower and
middle atmosphere by taking the newly measured rate
constants into account [17,18]. Our estimation is consis-
tent with those suggestions.
In this study, the room-temperature rate constant
for O(1D) + N2O reaction has also been determined
to be kN O ¼ ð1.35 ꢀ 0.08Þ ꢁ 10ꢂ10 cm3 moleculeꢂ1 sꢂ1
,
2
as listed in Table 1. Our kN O value is roughly 20% larger
2
than the NASA/JPL and IUPAC recommended values,
although our value is still within the error limits of the
recommendations. The largest value among the previ-
ously determined values for O(1D) + N2O reaction was
reported to be (1.27 0.08) · 10ꢂ10 cm3 moleculeꢂ1 sꢂ1
by Dunlea and Ravishankara [17], which is close to
our value. They measured the kN O value in the temper-
ature range between 220 and 370 K, and observed no
2

200
K. Takahashi et al. / Chemical Physics Letters 410 (2005) 196–200
[7] R.F. Heidner III, D. Husain, J.R. Wiesenfeld, J. Chem. Soc.
Faraday Trans. II 69 (1973) 927.
significant temperature dependence as identified in some
previous studies. Their kN O values are systematically
higher than the current NASA/JPL recommendations
between 220 and 370 K.
2
[8] G.E. Streit, C.J. Howard, A.L. Schmeltekopf, J.A. Davidson,
H.I. Schiff, J. Chem. Phys. 65 (1976) 4761.
[9] J.A. Davidson, H.I. Schiff, G.E. Streit, J.R. McAfee, A.L.
Schmeltekopf, C.J. Howard, J. Chem. Phys. 67 (1977) 5021.
[10] S.T. Amimoto, A.P. Force, R.G. Gulotty Jr., J.R. Wiesenfeld, J.
Chem. Phys. 71 (1979) 3640.
Acknowledgements
[11] L.C. Lee, T.G. Slanger, Geophys. Res. Lett. 6 (1979) 165.
[12] P.H. Wine, A.R. Ravishankara, Chem. Phys. Lett. 77 (1981)
103.
The authors are grateful to Masahiro Kanada, Hide-
hiko Jindo and Noriji Toriyama (STEL) for their techni-
cal assistance. The authors are indebted to Prof.
Yoshizumi Kajii (Tokyo Metropolitan University) for
his advice on the water vapor measurements. This work
was supported by Grants-in-Aid from the Ministry of
Education, Culture, Sports, Science and Technology,
Japan. The research grant for Dynamics of the Sun–
Earth–Life Interactive System, No. G-4, the 21st Cen-
tury COE Program is also acknowledged. Y.M. thanks
to the Steel Industrial Foundation for the Advancement
of Environmental Protection Technology for financial
support.
[13] K.H. Gericke, F.J. Comes, Chem. Phys. Lett. 81 (1981) 218.
[14] J. Shi, J.R. Barker, Int. J. Chem. Kinet. 22 (1990) 1283.
[15] A.R. Ravishankara, E.J. Dunlea, M.A. Blitz, T.J. Dillon, D.E.
Heard, M.J. Pilling, R.S. Strekowski, J.M. Nicovich, P.H. Wine,
Geophys. Res. Lett. 29 (2002), doi:10.1029/2002GL014850.
[16] R.S. Strekowski, J.M. Nicovich, P.H. Wine, Phys. Chem. Chem.
Phys. 6 (2004) 2145.
[17] E.J. Dunlea, A.R. Ravishankara, Phys. Chem. Chem. Phys. 6
(2004) 2152.
[18] M.A. Blitz, T.J. Dillon, D.E. Heard, M.J. Pilling, I.D. Trought,
Phys. Chem. Chem. Phys. 6 (2004) 2162.
[19] Y. Matsumi, S.M. Shamsuddin, Y. Sato, M. Kawasaki, J. Chem.
Phys. 101 (1994) 9610.
[20] Y. Matsumi, A.M.S. Chowdhury, J. Chem. Phys. 104 (1996)
7036.
[21] K. Takahashi, Y. Matsumi, M. Kawasaki, J. Phys. Chem. 100
(1996) 4084.
References
[22] K. Takahashi, N. Taniguchi, Y. Matsumi, M. Kawasaki, M.N.R.
Ashfold, J. Chem. Phys. 108 (1998) 7161.
[1] H. Okabe, Photochemistry of Small Molecules, Wiley-Inter-
sciences, New York, 1978.
[23] N. Taniguchi, K. Takahashi, Y. Matsumi, S.M. Dylewski, J.D.
Geiser, P.L. Houston, J. Chem. Phys. 111 (1999) 6350.
[24] N. Taniguchi, K. Hirai, K. Takahashi, Y. Matsumi, J. Phys.
Chem. A104 (2000) 3894.
[2] B.J. Finlayson-Pitts, J.N. Pitts Jr., Chemistry of the Upper and
Lower Atmosphere, Academic Press, New York, 2000.
[3] A.R. Ravishankara, G. Hancock, M. Kawasaki, Y. Matsumi,
Science 280 (1998) 60.
[25] M. Kono, Y. Matsumi, J. Phys. Chem. A105 (2001) 65.
[26] K. Takahashi, N. Taniguchi, Y. Sato, Y. Matsumi, J. Geophys.
Res. 107 (2002), doi:10.1029/2001JD001270.
[4] Y. Matsumi, M. Kawasaki, Chem. Rev. 103 (2003) 4767.
[5] S.P. Sander, R.R. Friedl, D.M. Golden, M.J. Kurylo, R.E. Huie,
V.L. Orkin, G.K. Moortgat A.R. Ravishankara, C.E. Kolb, M.J.
Molina, B. Finlayson-Pitts, J. Chemical Kinetics and Photochem-
ical Data for Atmospheric Studies, JPL Publication 02-25, JPL,
Pasadena, CA, 2003.
[27] K. Takahashi, S. Hayashi, Y. Matsumi, N. Taniguchi, S.
Hayashida, J. Geophys. Res. 107 (2002) 4440, doi:10.1029/
2001JD002048.
[28] S. Nishida, F. Taketani, K. Takahashi, Y. Matsumi, J. Phys.
Chem. A 108 (2004) 2710.
[6] R. Atkinson, D.L. Baulch, R.A. Cox, C.J. Howard, R.F.
Hampson, J.A. Kerr, M.J. Rossi, J. Troe, J. Phys. Chem. Ref.
Data 26 (2001) 521.
[29] R. Hilbig, R. Wallenstein, Appl. Opt. 21 (1982) 913.
[30] R.F. Heidner III, D. Husain, Int. J. Chem. Kinet. 6 (1974)
77.