The direct production of CO(v=1-9) in the reaction of O(3P) with the ethyl radical

Article abstract of DOI:10.1063/1.1288791

A new product channel that yields vibrationally excited CO(υ=1-9) in the reaction of the ethyl radical with O(³P) is experimentally observed by time-resolved Fourier transform infrared emission spectroscopy. The branching ratios for the different vibrational states are estimated to be 0.21+/-0.06, 0.27+/-0.03, 0.14+/-0.02, 0.08+/-0.02, 0.07+/-0.02, 0.07+/-0.02, 0.06+/-0.02, 0.05+/-0.02, and 0.05+/-0.02 for υ=1-9, respectively. Previously, only the CH3+H2CO, CH3CHO+ H, and C2H4+OH channels were known. Kinetics tests are provided to verify that the CO is produced directly in the reaction and not from secondary chemistry. The two possible new product channels are CO+CH4+H and CO+CH3+H2. The implications of this previously unexplored reaction channel for combustion chemistry and the possible mechanisms for this reaction are discussed.

Full text of DOI:10.1063/1.1288791

The direct production of CO(v = 1–9) in the reaction of O(3P) with the ethyl
radical
Jonathan P. Reid, Timothy P. Marcy, Seppe Kuehn, and Stephen R. Leone
Citation: J. Chem. Phys. 113, 4572 (2000); doi: 10.1063/1.1288791
View online: http://dx.doi.org/10.1063/1.1288791
View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v113/i11
Published by the American Institute of Physics.
Additional information on J. Chem. Phys.
Journal Homepage: http://jcp.aip.org/
Journal Information: http://jcp.aip.org/about/about_the_journal
Top downloads: http://jcp.aip.org/features/most_downloaded
Information for Authors: http://jcp.aip.org/authors
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

JOURNAL OF CHEMICAL PHYSICS
VOLUME 113, NUMBER 11
15 SEPTEMBER 2000
3
The direct production of CO„vÄ1–9… in the reaction of O„ P…
with the ethyl radical
Jonathan P. Reid, Timothy P. Marcy, Seppe Kuehn,a) and Stephen R. Leoneb)
JILA, University of Colorado and National Institute of Standards and Technology, Department of Chemistry
and Biochemistry, University of Colorado, Boulder, Colorado 80309-0440
͑Received 7 June 2000; accepted 21 June 2000͒
A new product channel that yields vibrationally excited CO( ϭ1–9) in the reaction of the ethyl
v
radical with O(3P) is experimentally observed by time-resolved Fourier transform infrared emission
spectroscopy. The branching ratios for the different vibrational states are estimated to be 0.21Ϯ0.06,
0.27Ϯ0.03, 0.14Ϯ0.02, 0.08Ϯ0.02, 0.07Ϯ0.02, 0.07Ϯ0.02, 0.06Ϯ0.02, 0.05Ϯ0.02, and 0.05Ϯ0.02
for ϭ1–9, respectively. Previously, only the CH ϩH CO, CH CHOϩH, and C H ϩOH channels
v
3
2
3
2
4
were known. Kinetics tests are provided to verify that the CO is produced directly in the reaction
and not from secondary chemistry. The two possible new product channels are COϩCH4ϩH and
COϩCH3ϩH2 . The implications of this previously unexplored reaction channel for combustion
chemistry and the possible mechanisms for this reaction are discussed. © 2000 American Institute
of Physics. ͓S0021-9606͑00͒00935-1͔
I. INTRODUCTION
In recent room temperature flow tube studies of combus-
tion related reactions involving hydrocarbon radicals longer
than CH3 , vibrationally excited CO has been observed at a
higher than expected level.9 The result can be best explained
The reactions of alkyl radicals with atomic and molecu-
lar oxygen are of fundamental importance in the combustion
of hydrocarbons.1,2 At combustion temperatures, hydrocar-
bon radicals react predominantly with oxygen atoms, since
the equilibrium with molecular O2 ,
by including a channel for the direct production of CO( )
v
from reaction of O with the longer chain hydrocarbons.
However, the existence of this reaction channel for produc-
ing CO directly from longer chain hydrocarbons has yet to
be confirmed.
RϩO2ꢀRO2 ,
͑1͒
favors the uncombined reactants.2 It was recently established
by infrared emission that vibrationally excited CO is pro-
duced directly in the reaction of ground state oxygen atoms
O(3P) with the methyl radical CH .3 Together with the
Competing channels in the reaction of O(3P) with the
ethyl radical C H , which yield OHϩC H , CH ϩCH O,
͓
͔
2
5
2
4
3
2
and CH3CHOϩH, have been studied previously,8,10–12 al-
though the possibility of a CO producing channel has not
been explored. The set of reaction channels includes
͓
͔
͓
͔
3
competing product channel leading to the formation of form-
aldehyde, these two channels,
3
C H ϩO P͒→H COϩCH
⌬HϭϪ318.6 kJ/mol, ͑3a͒
͑
2
5
2
3
3
CH ϩO P͒→H COϩH ⌬HϭϪ286 kJ/mol, ͑2a͒
͑
3
2
→CH3CHOϩH ⌬HϭϪ304.4 kJ/mol,
3
CH ϩO P͒→COϩH ϩH ⌬HϭϪ287 kJ/mol,
͑
3
2
͑3b͒
͑3c͒
͑2b͒
→C2H4ϩOH ⌬HϭϪ264.5 kJ/mol,
→CH3CHϩOH ⌬Hϭϩ150.7 kJ/mol,
are considered to be the dominant pathways.3–6 From the
infrared emission experiment, the branching ratio of channel
͑2b͒ was estimated to be as large as 40%.3 Recent experi-
ments by Fockenberg et al. determined a branching ratio for
this channel of 0.17Ϯ0.11 by time-of-flight mass spectrom-
etry and 0.18Ϯ0.04 by infrared diode laser spectroscopy.5
The formaldehyde channel was shown to account for ϳ0.8
of the products,5 in agreement with another estimate for
channel ͑2a͒ made by Niki et al.7 Hoyermanns and Sievert
concluded that the HCO product channel,
͑3d͒
͑3e͒
͑3f͒
→COϩCH4ϩH ⌬HϭϪ346.0 kJ/mol,
→COϩCH3ϩH2 ⌬HϭϪ343.0 kJ/mol.
A total rate constant of 2.2Ϯ0.4ϫ10Ϫ10cm3 moleculeϪ1 sϪ1
is reported by Slagle et al. and branching ratios are estimated
to be 0.32Ϯ0.07, 0.40Ϯ0.04, and 0.23Ϯ0.07 for channels
3͑a͒, 3͑b͒, and 3͑c͒, respectively.10 The first two reaction
channels are postulated to proceed through an excited ethoxy
3
CH ϩO P͒→HCOϩH
⌬HϭϪ354.2 kJ/mol,
͑
3
2
͑2c͒
has a maximum branching ratio of 0.2.8
*
radical intermediate, C2H5O , formed by the association of
O(3P) with C2H5 , with a well depth of 390 kJ molϪ1 12 The
subsequent products are formed by the decomposition of this
energized intermediate. In contrast, OH production is
.
a͒
NSF-REU program.
Staff member, Quantum Physics Division, National Institute of Standards
and Technology.
b͒
0021-9606/2000/113(11)/4572/9/$17.00
4572
© 2000 American Institute of Physics
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
CO(vϭ1–9) from O(3P) and ethyl radical
4573
thought to occur by direct abstraction of an H-atom, leading
to an inversion in the nascent vibrational product state dis-
frared emission is recorded at a fixed mirror displacement
with a liquid N2 cooled InSb detector. The laser is pulsed up
to 500 times at each mirror position with a repetition rate of
up to 100 Hz and the time dependence of the modulated
emission is recorded with a 16 bit analog-to-digital con-
verter. After the laser has pulsed the requisite number of
times for the number of coadditions ͑500͒, the interferometer
steps to the next mirror position and the procedure is re-
peated until a complete set of time-resolved interferograms
are acquired. The modulated emission is typically sampled
every 2 ␮s for a period of up to 1 ms, resulting in a total of
500 time slices. At the highest spectral resolution utilized in
this work of 2 cmϪ1, the modulated emission is sampled at
15 000 mirror positions. After initial experiments are per-
formed to probe the entire spectral range accessed by the
InSb detector ͑1700–8000 cmϪ1͒, subsequent experiments
are performed with a reduced Nyquist frequency by using a
bandpass interference filter to exclude emission outside the
spectral range 1700–5200 cmϪ1. This reduces the number of
interferogram points at which the modulated emission must
be sampled to 5212.
To ensure that the time-dependent histories of the prod-
uct species are not limited by the bandwidth of the amplifi-
cation electronics and detector rise time, the combined rise
time of the detector and amplifier was determined to be ϳ2
␮s. The product emission was observed to rise over a time
scale of 20 ␮s, which is thus accurately probed by our mea-
surements. The electronic bandwidth of the amplification
was systematically reduced to filter out high-frequency noise
while ensuring that no distortion of the emission profiles
results.
The pressures of the SO2 and either ethyl iodide or di-
ethyl ketone precursors in the flow regions prior to their mix-
ing are recorded with capacitance manometer pressure
gauges. This enables reproducible experiments to be per-
formed with comparable reagent number densities. Prior to
an experiment, the ethyl iodide or diethyl ketone precursors
are degassed twice through a freeze–pump–thaw cycle. The
pressure of SO2 prior to the photolysis region is typically
260 Pa while that of the ethyl iodide or diethyl ketone is
typically 130 Pa.
Fast Fourier transformation of the time-resolved inter-
ferograms leads to a sequence of spectra at 2 ␮s time inter-
vals over a 1 ms time window. The spectra are corrected for
the instrument response function of the spectrometer with a
calibrated blackbody source. Any contribution to the spectra
from blackbody emission of the optics must be subtracted. In
the time-resolved experiments, applying a high-pass filter re-
moves this DC signal component. Prior to the laser pulse,
ϳ100 ␮s of baseline are acquired to ensure that no emission
remains from the previous laser pulse and that no contribu-
tion from blackbody emission is observed. An example of
the spectra obtained at a spectral resolution of 5 cmϪ1 is
illustrated in Fig. 1, with product emissions identified from
the C–H stretch of the aldehydes CH3CHO and H2CO, and
from OH, CO, and SO.
tribution, which peaks for OH( ϭ3) and extends up to
v
OH( ϭ5).11
v
Following our earlier studies to measure the OH( )
v
product state distribution,11 we now examine the O(3P)
ϩC2H5 reaction again by time-resolved Fourier transform
infrared emission spectroscopy with the specific motivation
of exploring the possibility of a direct CO producing chan-
nel. The earlier experiments were restricted to the OH emis-
sion wavelengths because of interfering CO( ) that is pro-
v
duced in the photolysis of the C2H5 precursor, diethyl
ketone. Vibrationally excited CO( ϭ1–9) is now observed
v
from the O(3P)ϩC2H5 reaction and comprehensive tests are
performed to verify that the product channel is a conse-
quence of the direct reaction between O(3P) and C2H5 . In
Sec. II the experimental method is discussed. Section III pre-
sents the experimental data along with the comprehensive
tests that were performed to determine the source of the CO
product. In Sec. IV, we discuss the implications of this new
channel in the oxidation of hydrocarbon radicals and the pos-
sible mechanisms.
II. EXPERIMENTAL METHOD
Time-resolved Fourier transform infrared ͑TR-FTIR͒
emission spectroscopy is a powerful tool for probing the vi-
brationally excited products of reactions.13–16 Not only can
the emission from multiple product species be probed simul-
taneously, but the time-resolved history of these products
enables their origin to be postulated. We recently imple-
mented a commercial step-scan interferometer in our labora-
tory for the acquisition of time-resolved interferograms from
vibrationally excited molecules.17 For detection in the infra-
red we can acquire a time-resolved history of the emitting
molecules with a time resolution of ϳ2 ␮s and a spectral
resolution of 0.11 cmϪ1. Details of the experimental tech-
nique are given in our previous publications and only a con-
cise description is presented here.17
The reaction is initiated by simultaneous dissociation of
SO2 and either diethyl ketone or ethyl iodide by an ArF
193.3 nm excimer laser with an energy of ϳ100 mJ/pulse,
triggered by the FTIR spectrometer after the moving mirror
of the interferometer has been stabilized at an interferogram
position. The laser pulse permits the time-resolved history of
the product species to be attained by defining a reaction ini-
tiation time, tϭ0 ␮s. Cylindrical lenses are used to bring the
ArF laser beam to a collimated profile of 5 mmϫ5 mm in the
reagent flow region. The reagents flow into a reaction cham-
ber and are mixed prior to the photodissociation region. The
reaction chamber is evacuated and a 300 L/s roots blower
pump, backed by an 80 L/s mechanical pump, maintains
steady flow conditions. Simultaneous photodissociation of
the precursor species SO2 and either ethyl iodide or diethyl
ketone produces the reagents O(3P) and C2H5 . Reaction by
channels 3͑a͒–3͑f͒ leads to vibrationally excited products,
which emit in the infrared.
The infrared emission is collected by Welsh cell optics
and imaged into the FTIR spectrometer with a telescope that
matches the f/4.5 optics of the spectrometer. Modulated in-
The experiment was performed both without and with
the SO precursor, in order to determine the CO( ) yields
v
2
produced in the photodissociation of diethyl ketone and from
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

4574
J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
Reid et al.
FIG. 1. Infrared emission spectrum from the reaction of O(3P) with C2H5 at
a spectral resolution of 5 cmϪ1 and corresponding to one time slice at 10 ␮s
after the laser pulse initiates the reaction ͑
͒ corresponds to the photodissociation of diethyl ketone alone and
shows the CO( ) and the vibrationally excited C H radical products, also
͒. The lower spectrum
͑
v
2
5
at 10 ␮s after the laser initiation.
the reaction, respectively. Concentrations of the ethyl radical
and oxygen atom are similar magnitudes in the reaction ex-
periments. Experiments performed to test the dependence of
the reaction rates and product yields on the relative reagent
flow rates, summarized in Fig. 2, show that neither reagent is
in such an excess that pseudofirst-order kinetics with respect
to either the oxygen atom or ethyl radical is observed. The
integrated band intensity at the peak maximum in time of the
product signal provides a measure of the reaction rate rela-
tive to the relaxation rate of the vibrationally excited prod-
ucts, which is approximately constant from experiment to
experiment. The absorption cross sections of SO2 , diethyl
ketone, and ethyl iodide at 193 nm are 6ϫ10Ϫ18, 2.8
ϫ10Ϫ18, and 1.3ϫ10Ϫ18 cm2 moleculeϪ1, respectively.6,18,19
Thus, from the relative number densities of the different pre-
cursor species measured prior to the photolysis region, the
ratio of O(3P) to C2H5 is typically 4:1 with the diethyl ke-
tone precursor and 9:1 with the ethyl iodide precursor. All
subsequent experiments were performed under conditions
corresponding to the relative flow rates of 1.6 for SO2 and
0.8 for diethyl ketone shown in Fig. 2, which corresponds to
the pressures referred to above of 260 Pa of SO2 and 130 Pa
of ethyl iodide or diethyl ketone. These conditions are such
that the product yields and production rates are linearly de-
pendent on each reagent.
FIG. 2. Dependence of the CO ͑᭿͒, OH ͑᭺͒, and aldehyde ͑᭝͒ yields on:
͑a͒ SO2 precursor concentration ͑130 Pa diethyl ketone͒; ͑b͒ Alkyl radical
precursor concentration ͑260 Pa SO2͒; ͑c͒ Laser power ͑260 Pa SO2 and 130
Pa diethyl ketone͒. Due to constraints imposed by the signal-to-noise ratio,
the relative yields of the product channels are estimated by the integrated
band intensities of the different product vibrational bands at a time delay of
30 ␮s after the laser initiates the reaction. This assumes that the relaxation
rate of each product is constant within the variation in the conditions shown.
III. RESULTS
The main purpose of this work is to present evidence for
A. Possible origins for the CO product
the direct production of CO( ) in the reaction of O(3P) with
v
The vibrationally excited CO product can originate from
a number of secondary or minor reactions as well as from the
proposed direct reaction. In addition to listing these possible
sources we will discuss the specific proof required for each if
it were indeed responsible for the CO product observed.
C H . Although the spectrum in Fig. 1 reveals that CO( ) is
v
2
5
indeed produced when oxygen atoms are formed together
with ethyl radicals, many possible sources of the vibra-
tionally excited CO must be considered. We will first con-
sider the possible origins of the CO product followed by the
evidence that strongly suggests that the origin is the direct
reaction of O(3P) with C2H5 , rather than from a secondary
reaction source.
3
1. Direct product from the O„ P…¿C2H5 reaction
Following formation of the energized ethoxy radical in-
termediate, possible decomposition channels could be the
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
CO(vϭ1–9) from O(3P) and ethyl radical
4575
of two pathways. Oxygen atoms can react with the methyl
radical produced from the reaction ͑see Ref. 20 and refer-
ences therein͒
loss of CH4ϩH or CH3ϩH2 coupled in each case with the
formation of CO. The exothermicity of these reactions would
permit the partitioning of vibrational excitation into the CO
product, with a maximum vibrational quantum of 14 being
energetically accessible for the O(3P)ϩC2H5 reaction. This
is analogous to the reaction of methyl radicals with O(3P),
3
C H ϩO P͒→CH ϩHCO ⌬HϭϪ117.0 kJ/mol,
͑
2
4
3
͑4͒
to yield CO. Alternatively, CO can be produced from the
sequence of abstraction of H from C2H4 with a partial rate
constant21 of only ϳ2.4ϫ10Ϫ15 cm3 moleculeϪ1 sϪ1 fol-
lowed by reaction of the vinyl radical with an oxygen
atom,22
which leads to the observation of CO( ϭ0–8) and a calcu-
v
lated maximum accessible vibrational state of ϭ12.3 If the
v
direct channel is the source of the observed CO, the yield of
the CO( ) product must be linearly dependent on the con-
v
centration of ethyl radicals, i.e., linear with number density
of ethyl iodide or diethyl ketone, in the photolysis region.
When the conditions are chosen such that the O(3P) concen-
tration is not in excess, the CO yield must also be linear with
O(3P) concentration, i.e., linear with SO2 number density.
Finally, the CO yield must have a quadratic dependence on
laser power when O(3P) and C2H5 have similar concentra-
tions, since each CO results from the reaction of two pho-
tolysis products. The time-resolved appearance of the CO
product must also have a similar rise time to the appearance
of the other products that are known to originate from the
3
C H ϩO P͒→C H ϩOH,
͑5a͒
͑
2
4
2
3
3
C H ϩO P͒→CH ϩCO.
͑5b͒
͑
2
3
3
The low total reaction rate constant for the reaction O(3P)
ϩC2H4 ͑0.5–1ϫ10Ϫ12 cm3 moleculeϪ1 sϪ1) coupled with the
low branching ratio10 of Ͻ0.23 for production of C2H4 ͓Eq.
3͑c͔͒ could lead only to a low yield of CO and a slow rise
time relative to that from the direct C2H5ϩO(3P) reaction,
which
has
a
total
rate
constant
of
ϳ2
10
ϫ10Ϫ10 cm3 moleculeϪ1 sϪ1
.
In addition, the CO yield
direct reaction, such as OH( ) and CH CHO( ). All of
v
v
would be either quadratic or cubic in its dependence on the
3
concentration of O(3P).
these processes, Eqs. 3͑a͒–3͑f͒ result in the depletion of
O(3P) and C2H5 and must thus reach a maximum product
concentration at an identical time, after one of the reactants
has been fully depleted. In addition, we might expect the
relative yield of CO to provide an indication of its source
when compared with the yield from the separately studied
reaction of CH3 with O(3P). This last point will be consid-
ered again in the discussion below.
4. Minor product channel and photolysis precursor
reactions
Secondary reactions with the products from channels
3͑a͒ and 3͑c͒ have been considered. Channel 3͑b͒ produces
CH3CHO which has
a total rate constant of ϳ5
ϫ10Ϫ13 cm3 moleculeϪ1 sϪ1 for reaction with O(3P), with
the dominant channel being the production of OH.23 Simi-
larly, the reaction of O(3P) with H2CO has a total rate con-
3
2. Secondary product from the O„ P…¿CH3 reaction
stant of ϳ1ϫ10Ϫ13 cm3 moleculeϪ1 sϪ1 24 Clearly, any other
.
Reaction channel 3͑a͒ for O(3P)ϩC2H5 leads to the pro-
duction of CH3 . This product could in turn lead to the sec-
ondary reaction of methyl radicals with oxygen atoms to
produce CO by the route that has now been corroborated by
several different experimental methods.3–5 Consider two hy-
pothetical experiments with comparable experimental condi-
tions: one experiment probes the O(3P)ϩC2H5 reaction
while the other probes the O(3P)ϩCH3 reaction. If the num-
ber densities of C2H5 and CH3 in the two experiments are
adjusted to be comparable, the yield of CO from this second-
ary reaction channel of C2H5 would be Ͻ30% of that from
the direct reaction experiment of O(3P) with CH3 . In addi-
tion, we would expect the CO signal from the secondary
mechanisms for the secondary production of CO would re-
sult in a delayed rise of CO relative to the initial products
formed in the reaction of C2H5 with O(3P).
At the laser fluence used in these studies, the only sig-
nificant hydrocarbon product expected in the photodissocia-
tion of diethyl ketone and ethyl iodide is the ethyl
radical.18,25 Diethyl ketone has been used previously as a
precursor for the ethyl radical in studies of the reaction of
C2H5 with Cl(2P) ͑Ref. 26͒ and O(3P).9,10 To our knowl-
edge, no detailed studies have been performed of the branch-
ing ratios for competing bond cleavages in the photolysis of
diethyl ketone. However, we expect that the dissociation of
diethyl ketone at 193 nm will be analogous to the dissocia-
tion of acetone. In studies of the fragmentation of acetone27
the quantum efficiency of producing CO and methyl radicals
is reported to be ϳ0.96. Two other channels,
reaction to be delayed in its rise-time relative to the OH( )
v
and aldehyde products from channels 3͑a͒–3͑c͒. As final
conclusive evidence, we would expect the magnitude of the
CO product signal to be quadratic in its dependence on the
concentration of O(3P), with each CO product molecule re-
sulting from the reaction of C2H5 with O(3P) and followed
by CH3 with O(3P).
CH ͒ CO→CH COCH ϩH,
͑6a͒
͑6b͒
͑
͑
3 2
2
3
CH ͒ CO→CH COϩCH ,
3 2
2
4
are reported to contribute less than 4% to the product frag-
ments. A secondary minor photolysis product is the methyl-
ene radical, CH2 , produced from the photolysis of CH3 , and
this is known to produce CO in reaction with O(3P).27 How-
ever, the yield of the methylene radical is expected to be
Ͻ5% of the yield of the methyl radical in the photodissocia-
3
3. Secondary product from the O„ P…¿C2H4 reaction
Reaction channel 3͑c͒ leads to the production of C2H4
which can then react further with O(3P) to yield CO by one
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

4576
J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
Reid et al.
tion of acetone, under our experimental conditions.3 Simi-
larly, we expect the yield of analogous secondary photolysis
products to be much lower than the yield of ethyl radicals in
the photolysis of diethylketone and, thus, to contribute insig-
nificantly to the yield of CO observed in the reaction of
C2H5ϩO(3P).
We must, however, consider reaction with the undisso-
ciated precursor species. Reaction of O(3P) with C2H5I
leads to the production of C2H5ϩIO and C2H4ϩHOI.28 Both
of these hydrocarbon products are then able to react with
another O(3P) atom in the ways already described, with the
former contributing to the reactions 3͑a͒–3͑f͒ that are
probed, and the latter being much slower than the sum of
channels 3͑a͒–3͑f͒. Diethyl ketone is also able to react with
O(3P) but at only a very slow rate (kϳ1.7ϫ10Ϫ14
cm3 moleculeϪ1 sϪ1) and leading only to the OH product.29
The photodissociation of diethyl ketone leads directly to the
production of CO( ϭ0–2) and the contribution of these
v
product states to the CO product distribution can be easily
subtracted.18
3
FIG. 3. Time-evolution of: ͑a͒ product species from the reactions of O(3P)
B. Evidence for the direct reaction of O„ P… with C2H5
to yield the CO product
with C2H5 following the laser initiation: CO ͑
͒, OH ͑
͒, and
• • • •
aldehyde ͑
͒. Only CO( Ͼ6) are included in the integrated band
v
In the preceding section we established criteria that en-
able us to identify the origin of the CO produced when
O(3P) and C2H5 are simultaneously prepared by laser initia-
tion. We now present the evidence that identifies the direct
reaction of O(3P) with C2H5 as the source of the observed
CO.
intensity of the CO signal to avoid complications from vibrational cascading
effects. ͑b͒ CO signal from the photodissociation of acetone ͑
͒ and
3
• • • •
diethyl ketone ͑
͒ and from the reactions of O( P) with C2H5 ͑
͒
•
•
•
and CH3
͑
͒. The photodissociation products have a prompt rise
while the reaction products are delayed. The longer time decay of the CO
signal in panel ͑b͒ compared with panel ͑a͒ illustrates the slower relaxation
rate of CO( ϭ1–2) produced in the photodissociation than the relaxation
v
of the high vibrational states ( Ͼ6) produced in the reaction. Flow condi-
tions: 260 Pa SO2 and 130 Pa diethyl ketone.
v
1. Dependence on reagent number densities
The yield dependencies of the CO( ), OH( ), and com-
v
v
bined C–H stretching bands of the aldehydes H CO( ) and
v
2
confirmed by the quadratic dependence of the CO yield on
laser power, as shown in Fig. 2, taken under conditions such
that the O atom concentration is similar to the C2H5 concen-
tration.
CH CHO( ) on the relative number densities of the reagents
v
3
are presented in Fig. 2. In each case, one reagent flow rate is
held constant while the other is varied. The relative OH( )
v
and CO( ) yields are calculated by integrating the intensities
v
of the vibrational bands originating from OH( ϭ3) and
v
2. Reaction rise-time of product species
CO( ϭ6–12) at a time delay corresponding to their maxi-
v
mum integrated emission intensity of ϳ30 ␮s. These states
are chosen because they represent the high nascent vibra-
tional states produced by the reaction, at a time before sig-
nificant relaxation of the vibrational states complicates the
time dependence of the emission profiles. Linear dependen-
cies are observed with flow rate of SO2 , diethyl ketone, and
ethyl iodide ͑not shown͒. In each case, it can be concluded
that the yield of CO is linearly dependent on both the con-
centration of C2H5 and O(3P) and, thus, involves the reac-
tion of one C2H5 radical with one atom of O(3P) in the rate
limiting step. This is demonstrated for two different photoly-
The products from the three channels 3͑a͒–3͑c͒, produc-
ing H2CO and CH3CHO, denoted as the C–H aldehyde
stretching region in Fig. 1, and OH, all have similar rise
times within the signal-to-noise ratio of our experiments, and
these rise times are comparable to the rise time of the high
vibrational states of the CO product. All have time constants
of ϳ10 ␮s, as shown in Fig. 3͑a͒. Comparison cannot be
made with the low vibrational states of the CO product since
these states have a complicated time dependence due to the
vibrational cascading by relaxation of higher vibrational
states.3 From the analysis above, we find that the total reac-
tion rate is much larger than would be expected from sec-
ondary reactions and this is attributed to the larger number
densities of the reactants in the direct reaction of C2H5 with
O(3P) than the reactants involved in any secondary chemis-
try.
sis precursors that yield the C2H5 reactant. If the CO( )
v
were formed by a secondary reaction, the secondary step
would become the rate limiting step in forming CO and the
dependence of the CO product yield on the concentration of
C2H5 and O(3P) would become more complex. The similar
linear dependencies in yields of the OH, the aldehydes, and
CO products with reagent number densities strongly suggest
that the direct reaction of C2H5 and O(3P) is the origin of
the observed CO, as well as of OH and the aldehydes. This is
Experiments are also performed to compare the CO rise
times from the reaction of C2H5 with O(3P) and CH3 with
O(3P) using similar number densities of the alkyl radicals
and oxygen atoms. The results of these comparative studies
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
CO(vϭ1–9) from O(3P) and ethyl radical
4577
TABLE I. Approximate vibrational distributions of CO( ) produced in the
v
4. CO product state distribution
reactions of O(3P) with C2H5 and of O(3P) with CH3 at 20 ␮s following
The product state distribution given in Table I is com-
laser initiation of the reaction. Overlap of emission bands from SO( ) at
early times and vibration relaxation preclude an accurate estimate of the
v
pared with that from the reaction of O(3P) and CH3 . The P
nascent population in CO( ϭ1) and thus an accurate estimate of the na-
v
and R branches of the vibrational emission from ϭ1–12
v
scent distribution. Populations quoted for CO( ϭ1) for the current study
v
were included in a least-squares fit to the vibrational band
contour, using the known Einstein A coefficients32 and cal-
culating the rovibrational transition energies from the pub-
lished spectroscopic constants of CO.33 The nascent vibra-
tional distributions of the CO product are similar from the
reaction of O(3P) with CH3 and from O(3P) with C2H5 .
are determined from the Boltzmann temperatures calculated from CO(
v
ϭ2–9). The sum of the fractional populations for CO( ϭ1–9) is normal-
v
ized to 1 and the errors are determined from the spectral baseline noise
͑Ϯ1␴͒.
O(3P)ϩCH3
CO vibrational state O(3P)ϩC2H5
This study
Seakins and Leonea
Overlap between emission from SO( ) and CO( ϭ1) pre-
v
v
cludes an accurate spectroscopic estimate of the population
1
2
3
4
5
6
7
8
9
͑0.21Ϯ0.06͒ ͑0.20 Ϯ 0.13͒
0.27 Ϯ 0.04
0.19 Ϯ 0.03
0.15 Ϯ 0.02
0.12 Ϯ 0.02
0.08 Ϯ 0.01
0.06 Ϯ 0.01
0.08 Ϯ 0.02
0.05 Ϯ 0.01
¯
0.27Ϯ0.03
0.14Ϯ0.02
0.08Ϯ0.02
0.07Ϯ0.02
0.07Ϯ0.02
0.06Ϯ0.02
0.05Ϯ0.02
0.05Ϯ0.02
0.33 Ϯ 0.03
0.14 Ϯ 0.02
0.07 Ϯ 0.02
0.07 Ϯ 0.02
0.06 Ϯ 0.02
0.05 Ϯ 0.02
0.05 Ϯ 0.02
0.03 Ϯ 0.02
in CO( ϭ1). The population values quoted in Table I for
v
CO( ϭ1) represent those values estimated from a Boltz-
v
mann fit to the vibrational distribution, where ϭ1 is on the
v
fit line and ϭ2 is higher than the fit, possibly due to relax-
v
ation. In addition, the population in the high vibrational
states ( Ͼ9) of CO was observed to be Ͻ10%. An accurate
v
estimate of the population in these states could not be ob-
tained due to the rapid decline in sensitivity of the InSb
detector at frequencies below 1850 cmϪ1. With a different
detector it would be possible to examine these high vibra-
tional states in more detail, along with the CO stretching
vibrational mode of the aldehyde products at around 1750
aReference 3.
are shown in Fig. 3͑b͒, along with the relative yields of CO
produced from the photodissociation of the two parent spe-
cies, diethyl ketone and acetone. The rise time to maximum
signal is marginally shorter for the reaction of C2H5 with
O(3P) than for reaction with CH3 . This is in agreement with
the total rate constants for these two reactions, which are
ϳ2.2ϫ10Ϫ10 cm3 moleculeϪ1 sϪ1 and ϳ1.7ϫ10Ϫ10 cm3
moleculeϪ1 sϪ1, respectively,5,10 leading to a more rapid
depletion of the reactants in the former case. The oxygen
atom concentration is in excess of the alkyl radical concen-
tration, and thus the alkyl radical concentration limits the
final total yield of product species.
cmϪ1. The CO( ϭ2–9) distribution from the reaction of
v
O(3P) with C2H5 can be fit to a Boltzmann temperature of
14 500Ϯ2500 K compared to 14 400Ϯ2400 K for the O(3P)
with CH3 reaction. The latter of these two temperatures is in
good agreement with the temperature determined by Seakins
and Leone of 12 700Ϯ1400 K.3
5. Kinetic modeling
A kinetic model for the reaction system was used to
investigate the relative production rates of the CO product
from different reaction channels. From the known rate con-
stants and branching ratios previously discussed and refer-
enced in this work, reaction channels 2͑a͒, 2͑b͒, and 3͑a͒–
3͑c͒, 3͑e͒–3͑f͒ were included in the kinetic model. The
combined rate constant for channels 3͑e͒ and 3͑f͒ was chosen
to be equal to that for channel 2͑b͒. The rate of depletion of
O(3P) by reaction with C2H4 was also included. The initial
experimental rise time of the OH product from reaction 3͑c͒
was used to estimate the approximate absolute number den-
sities of C2H5 and O(3P) produced in the photodissociation
of the precursors. Once these initial concentrations were es-
timated, the rate of production of CO from secondary reac-
tions involving CH3 , by channels 3͑a͒ and 2͑b͒, and C2H4 by
channels 3͑c͒ and ͑4͒, was found to be insufficient to explain
the experimentally measured rise time of the CO product. In
Fig. 4, the initial fit of the OH signal rise is shown, along
3. Total CO yield
When the yields of CO from the reactions of C2H5 with
O(3P) and CH3 with O(3P) are compared under similar ex-
perimental conditions, the yield of CO( ) in the former case
v
is ϳ0.9 of that in the reaction of O(3P) and CH3 at 20 ␮s
after laser initiation. To estimate this yield we assume that
the relative integrated CO signals from the photodissociation
of acetone and diethyl ketone provide a measure of the rela-
tive yields for producing CH3 and C2H5 from the two pre-
cursor species ͑1:0.7͒. The absorption cross-sections at 193
nm for acetone and diethyl ketone are 3.8ϫ10Ϫ18 and 2.8
ϫ10Ϫ18 cm2 moleculeϪ1, respectively,18,30,31 in approximate
agreement with the estimate of the relative CO( ) yields
v
from the photodissociation. An assumption is also made in
estimating the relative yields that the vibrational distributions
for the reactions of C2H5 with O(3P) and CH3 with O(3P)
are similar. The approximate vibrational distributions are
measured and Table I shows that this is a reasonable assump-
tion. This supports the conclusion that the CO must be pro-
duced from the direct reaction as it cannot be produced in
such a substantial yield from secondary reaction mecha-
nisms.
with the relative contributions of the CO( ) produced by
v
several competing channels. It can be seen that the secondary
reaction of CH3 with O(3P), where the CH3 is generated by
reactions 3͑a͒, 3͑f͒ and ͑4͒, is a significant contributor to a
long time rise of the CO product channel. Such a slow rise is
convoluted with the vibrational deactivation of CO. How-
ever, the rapid initial rise time of the CO( ) product can
v
only be explained by the direct reaction of C2H5 and O(3P).
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

4578
J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
Reid et al.
models confirm that the rapid rise in the CO( ) signal and
v
the CO( ) yield cannot result from secondary chemistry.
v
Previous studies in this laboratory conclusively showed
that CO is produced directly in the reaction of CH3 with
O(3P).3 Additional tests were performed that are not re-
peated here, which confirm that the reaction proceeds with
O(3P) and not by reaction with electronically excited
O(1D). In addition, possible vibration-to-vibration energy
transfer from vibrationally excited SO, produced in the pho-
todissociation of SO2 , to CO, produced in the dissociation of
acetone, was excluded as a source for the vibrationally ex-
cited CO. The buffer gas was varied in both pressure and
identity ͑Ar, N2 , and He͒ to probe the influence of methyl
radical vibrational excitation on the CO product. Although,
the lifetimes of the ␯ and ␯ modes of the CH3 reactant
2
3
were made considerably shorter than the reaction time, no
change in the CO product distribution and yield was
observed.3 This eliminated the possibility that vibrational ex-
citation in the hydrocarbon radical significantly influences
the CO reaction channel. Fig. 1 shows that the ethyl radical
is born with considerable C–H stretching excitation follow-
ing the photodissociation of diethyl ketone. However, the
lifetime of this vibrational excitation is also considerably
shorter than the reaction time and, in addition, these studies
have been performed with two different precursor molecules
͑ethyl iodide and diethyl ketone͒, suggesting once again that
excitation in the alkyl radical has little impact on the reaction
dynamics.
FIG. 4. Kinetic model of the competing sources for the production of CO.
͑a͒ Time-evolution of the OH( ) product signal experiment;
kinetic fit͒. The experimental signal decreases at long times due to
deactivation of the high vibrational states of OH. The theoretical curve
͑
v
In a recent study of the reaction of O(3P) with C2H5 by
Hoyermann et al.,12 only reaction channels 3͑a͒–3͑c͒ were
probed by electron impact ionization. The possibility of a
CO producing channel was not examined. However, at the
electron energies necessary for probing the CO product chan-
nel, fragmentation of H2CO can occur, obscuring any CO
product that may be formed.
represents the total OH yield. ͑b͒ Time-evolution of the CO( ) product
v
signal ͑
experiment;
total kinetic fit of all reaction channels;
CO yield from competing channels as labeled͒. The early time rise of
the CO yield is underestimated by the model since some prompt CO is also
produced by the photodissociation of the diethyl ketone precursor.
It should also be noted that one of the products of this direct
CO producing reaction is CH3 , by reaction 3͑f͒, which can
then generate further CO. Thus, it is possible for the reaction
of C2H5 with O(3P) to yield two molecules of CO.
The mechanism for the CO product channel in the reac-
tion of O(3P) with C2H5 probably originates with the forma-
*
tion of the energized ethoxy radical, C2H5O , with a well
depth of 390 kJ molϪ1 12
.
Thus, not only channels 3͑a͒ and
IV. DISCUSSION
3͑b͒ proceed initially by this complex forming pathway, but
channels 3͑e͒ and 3͑f͒ as well.
Once the ethoxy radical is formed, a possible mechanism
for CO production could involve the elimination of either
CH3 or H leading to energetically excited H2CO or
CH3CHO, respectively, followed by the elimination of H2 or
The direct reaction channel of C2H5 with O(3P) to pro-
duce CO has not previously been explored. We have now
confirmed by time-resolved Fourier transform infrared emis-
sion spectroscopy the existence of a CO product channel
from this reaction and examined the nascent vibrational
product state distribution. By performing specific tests de-
signed to fulfill predetermined criteria we have confirmed
that the CO product channel does originate from the direct
reaction of C2H5 and O(3P). The dependence of the CO
yield on the reagent number densities and laser power con-
firms that one C2H5 radical and one atom of O(3P) are in-
volved in the rate determining step. A similarity in the rise
times of the products from the competing channels 3͑a͒–3͑f͒,
CH3CHO, H2CO, OH, and CO, strongly suggests that the
four products have an identical origin, namely the direct re-
action of C2H5 with O(3P). Comparable yields of CO from
the reactions of O(3P) with C2H5 and with CH3 suggest that
secondary chemistry cannot be the source of the observed
CO from the reaction of O(3P) with C2H5 . Indeed, kinetics
CH to yield CO( ). This is analogous to one mechanism
v
4
postulated for the CO product channel in the reaction of
O(3P) with CH3 .3 However, in that case Seakins and Leone
argued that the reaction could not proceed first by the elimi-
nation of H to form formaldehyde, followed by the elimina-
tion of H2 to form CO. A significant barrier exists for the
dissociation of formaldehyde to form CO.34 The barrier
height is reported to be ϳ350 kJ/mol compared to the exo-
thermicity of reaction 2͑a͒ which is only ϳ286 kJ/mol. Thus,
the energized H2CO product does not have enough energy to
surmount the barrier to form the CO product by reaction
pathway ͑7͒:
*
*
CH3O →H2CO ϩHCOϩH2ϩH.
͑7͒
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
CO(vϭ1–9) from O(3P) and ethyl radical
4579
The reactions producing formaldehyde and acetaldehyde by
channels 3͑a͒ and 3͑b͒ from the reaction of O(3P) with C2H5
are significantly more exothermic than channel 2͑a͒ in the
O(3P) with CH3 reaction, but the barrier to elimination of
These steps are analogous to reaction steps ͑11͒ and ͑12͒,
respectively.
The monotonically decreasing nascent CO vibrational
distribution suggests that the reaction does proceed through a
complex, supporting the ideas postulated above.37–40 How-
ever, as for the reaction of O(3P) with CH3 , it is difficult to
envisage any direct mechanism which could lead to the pro-
H or CH to yield CO( ) remains significantly higher than
v
2
4
the energy of the reactants, prohibiting channels analogous to
reaction ͑7͒ from proceeding.
Examining reaction pathway ͑7͒, recent theoretical cal-
culations by Harding and Klippenstein35 concluded that the
CO product could be formed dynamically following the
failed loss of an H-atom. They postulate that following the
extension of a C–H bond, the H-atom is prevented from
departing by a large centrifugal barrier. However, the depart-
ing H-atom can abstract a second H-atom to form H2 , then
readily separate to the H2 and HCO products over a much
lower barrier, followed by the subsequent decomposition of
HCO to HϩCO.
This is similar to a mechanism proposed by Seakins and
Leone3 in which the reaction of O(3P) with CH3 could pro-
ceed first by the elimination of H2 through a transition state
involving a concerted elimination of H2 , followed then by
the elimination of H from HCO. Such a transition state has
not been found in theoretical calculations.35 The first step
producing HCOϩH2 ,
duction of CO( ).
v
Further studies to probe longer chain hydrocarbon radi-
cals in this laboratory ͑neo-pentyl and iso-butyl͒ also lead to
a preliminary conclusion that the direct product of CO may
be a general channel in the reaction of hydrocarbon radicals
with O(3P). Ongoing work is attempting to quantify the
branching ratios and yields of these reactions.
Although the CO branching ratio may be lower than the
value of 0.4 initially estimated by infrared emission for these
reactions, even the lower amended value of 0.18 estimated
by Fockenberg et al.5 for the reaction of O(3P) with CH3
suggests that a significant channel may have previously been
omitted from models of the combustion of hydrocarbons.
ACKNOWLEDGMENTS
The authors thank the Department of Energy for support
of this research, and the National Science Foundation and the
Air Force Office of Scientific Research for additional equip-
ment.
*
*
CH3O →HCO ϩH2→COϩH2ϩH,
͑8͒
is exothermic by 359 kJ/mol. In an analogous way, CH4 or
H2 could be eliminated directly from the ethoxy radical and
this could then be followed by the elimination of H or CH3 ,
respectively, to form CO:
1 J. A. Miller, R. J. Kee, and C. K. Westbrook, Annu. Rev. Phys. Chem. 41,
345 ͑1990͒.
2 J. Warnatz, in Combustion Chemistry, edited by J. C. Gardiner ͑Springer-
Verlag, New York, 1993͒.
*
*
C2H5O →CHO ϩCH4→COϩCH4ϩH,
͑9͒
3 P. W. Seakins and S. R. Leone, J. Phys. Chem. 96, 4478 ͑1992͒.
4 Z. Min, R. W. Quandt, T.-H. Wong, and R. Bersohn, J. Chem. Phys. 111,
7369 ͑1999͒.
*
*
C2H5O →CH3CO ϩH2→COϩCH3ϩH2 .
͑10͒
A further mechanism for the reaction of O(3P) with
CH3 , proposed by Wagner ͑see details in Ref. 3͒, first in-
5 C. Fockenberg, G. E. Hall, J. M. Preses, T. J. Sears, and J. T. Muckerman,
J. Phys. Chem. A 103, 5722 ͑1999͒; J. M. Preses and C. Fockenberg
͑private communication͒.
*
*
volves the isomerization of CH3O to form H2COH . Once
this intermediate is formed, the reaction can proceed either
by elimination of H to form the formaldehyde product or by
elimination of H2 followed by H to form ultimately the CO
product:
6 I. R. Slagle, D. Sarzynski, and D. Gutman, J. Phys. Chem. 91, 4375
͑1987͒.
7 H. Niki, E. E. Daby, and B. Weinstock, International Symposium on Com-
bustion ͑The Combustion Institute, 1969͒, Vol. 12, p. 277.
8 K. Hoyermann and R. Sievert, International Symposium on Combustion
͑The Combustion Institute, 1979͒, Vol. 17, p. 517.
9 P. F. Zittel ͑private communication͒.
*
*
CH3O →H2COH →H2COϩH,
͑11͒
10 I. R. Slagle, D. Sarzynski, D. Gutman, J. A. Miller, and C. F. Melius, J.
Chem. Soc., Faraday Trans. 84, 491 ͑1988͒.
*
*
*
CH3O →H2COH →COH ϩH2→COϩHϩH2 .
͑12͒
11 J. Lindner, R. A. Loomis, J. J. Klaassen, and S. R. Leone, J. Chem. Phys.
108, 1944 ͑1998͒.
The reaction of O(3P) with C2H5 could proceed by a similar
mechanism. Wagner et al. have shown theoretically36 that
the reaction of O2 with C2H5 proceeds by the isomerization
of the initial addition complex C2H5O2 to CH2CH2OOH fol-
lowed by decomposition to ethene and HO2 . Isomerization
of the initial ethoxy complex, C2H5O, formed from O(3P)
12 K. Hoyermann, M. Olzmann, J. Seeba, and B. Viskolcz, J. Phys. Chem. A
103, 5692 ͑1999͒.
13 J. J. Sloan, Advances in Spectroscopy ͑Wiley, New York, 1989͒, Vol. 18.
14 S. R. Leone, Acc. Chem. Res. 22, 139 ͑1989͒.
15 S. A. Rogers and S. R. Leone, Appl. Spectrosc. 47, 1430 ͑1993͒.
16 G. Hancock and D. E. Heard, in Advances in Photochemistry, edited by D.
H. Volman, G. S. Hammond, and D. C. Neckers ͑Wiley, New York,
1993͒, Vol. 18, pp. 1–65.
*
and C2H5 , could lead to the intermediate CH3CHOH .
17 J. P. Reid, C. X. W. Qian, and S. R. Leone, Phys. Chem. Chem. Phys. 2,
853 ͑2000͒.
Elimination of an H-atom would yield the acetaldehyde
product ͑13͒:
18 G. E. Hall, H. W. Metzler, J. T. Muckerman, J. M. Preses, and R. E.
Weston, Jr., J. Chem. Phys. 102, 6660 ͑1995͒.
*
*
C2H5O →CH3CHOH →CH3CHOϩH.
͑13͒
19 C. M. Roehl, J. B. Burkholder, G. K. Moortgat, A. R. Ravishankara, and
P. J. Crutzen, J. Geophys. Res. 102, 12819 ͑1997͒.
*
Alternatively, CH4 could be eliminated to yield COH and,
20 O. K. Abou-Zied and J. D. McDonald, J. Chem. Phys. 109, 1293 ͑1998͒.
21 K. Mahmud, P. Marshall, and A. Fontijn, J. Phys. Chem. 91, 1568 ͑1987͒.
22 D. J. Donaldson, I. V. Okuda, and J. J. Sloan, Chem. Phys. 193, 37 ͑1995͒.
23 D. L. Baulch, C. J. Cobos, R. A. Cox, P. Frank, G. Hayman, T. Just, J. A.
thus, CO ͑14͒:
*
*
*
C2H5O →CH3CHOH →COH ϩCH4→COϩHϩCH4 .
͑14͒
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

4580
J. Chem. Phys., Vol. 113, No. 11, 15 September 2000
Reid et al.
Kerr, T. Murrells, M. J. Pilling, J. Troe, R. W. Walker, and J. Warnatz, J.
Phys. Chem. Ref. Data 23, 847 ͑1994͒.
33 K. P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure
IV. Constants of Diatomic Molecules ͑Van Nostrand Reinhold, New York,
24 J. S. Chang and J. R. Barker, J. Phys. Chem. 83, 3059 ͑1979͒.
25 J. Pola, Y. Koga, and A. Ouchi, Tetrahedron 53, 3757 ͑1997͒.
26 P. W. Seakins, E. L. Woodbridge, and S. R. Leone, J. Phys. Chem. 97,
5633 ͑1993͒.
1979͒.
34 C. B. Moore and J. C. Weisshaar, Annu. Rev. Phys. Chem. 34, 525
͑1983͒.
35 L. B. Harding and S. J. Klippenstein ͑private communication͒.
36 A. F. Wagner, I. R. Slagle, D. Sarzynski, and D. J. Gutman, J. Phys.
Chem. 94, 1853 ͑1990͒.
27 P. D. Lightfoot, S. P. Kirwan, and M. J. Pilling, J. Phys. Chem. 92, 4938
͑1988͒.
28 R. A. Loomis, J. J. Klaassen, J. Lindner, P. G. Christopher, and S. R.
Leone, J. Chem. Phys. 106, 3934 ͑1997͒.
37 D. J. Donaldson and J. J. Sloan, J. Chem. Phys. 82, 1873 ͑1985͒.
38 U. Schwanke, H. Heydtmann, and J. J. Sloan, Chem. Phys. 132, 413
͑1989͒.
29 J. M. Roscoe, Can. J. Chem. 64, 1458 ͑1986͒.
30 M. Baba, H. Shinohara, N. Nishi, and N. Hirota, Chem. Phys. 83, 221
͑1984͒.
39 E. Arunan, S. J. Wategaonkar, and D. W. Setser, J. Phys. Chem. 95, 1539
͑1991͒.
31 H. Ito, Y. Nogata, and S. Matsuzaki, Bull. Chem. Soc. Jpn. 42, 2453
͑1969͒.
40 E. Arunan, R. Rengaragan, and D. W. Setser, Can. J. Chem. 72, 568
͑1994͒.
32 C. A. Wight and S. R. Leone, J. Chem. Phys. 78, 4875 ͑1983͒.
Downloaded 04 Oct 2012 to 171.67.34.69. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions