Kinetics of the CCO + NO and CCO + NO2 reactions

Article abstract of DOI:10.1021/jp0304125

The kinetics of the reaction of CCO radicals with NO and NO₂ were studied using time-resolved infrared diode laser absorption spectroscopy. The rate constants were determined to be k_CCO+NO = (5.36 ± 0.5) × 10^-11 and k

Full text of DOI:10.1021/jp0304125

74
J. Phys. Chem. A 2004, 108, 74-79
Kinetics of the CCO + NO and CCO + NO2 Reactions
W. David Thweatt, Mark A. Erickson, and John F. Hershberger*
Department of Chemistry, North Dakota State UniVersity, Fargo, North Dakota 58105
ReceiVed: March 31, 2003; In Final Form: September 16, 2003
The kinetics of the reaction of CCO radicals with NO and NO
diode laser absorption spectroscopy. The rate constants were determined to be kCCO+NO ) (5.36 ( 0.5) ×
and kCCO+NO2 ) (6.89 ( 0.5) × 10 cm molecule at 298 K. These rate constants have virtually
no temperature dependence over the range 298-573 K, with Arrhenius fits given by kCCO+NO ) (1.66 ( 0.5)
2
were studied using time-resolved infrared
-
11
-11
3
-1 -1
10
s
-
10
-11
×
10 exp[(-337 ( 100)/T] and kCCO+NO2 ) (8.51 ( 1.5) × 10 exp[(-63 ( 80)/T] (error bars represent
one standard deviation). Product channel measurements were performed on the CCO + NO reaction. After
consideration of secondary chemistry, we obtain the branching ratios φ(CN + CO ) ) 0.13 ( 0.05 and φ(CO
NCO) ) 0.87 ( 0.05 at 298 K. Ab initio quantum chemical calculations at the QCISD(T)/6-311G(2df,2pd)
level of theory were used to examine probable reaction pathways.
2
+
Introduction
b) is obtained from standard tables,11 except for ∆Hf of NCO,
12
which was obtained from recent measurements. Since channels
1c), (1d), and (1e) are endothermic, they cannot be significant
Reactions between small molecular radicals and nitrogen
oxides are crucial elementary steps in models of NOx formation
and removal in combustion systems. One such species of
interest in combustion kinetics is the CCO radical. This molecule
has been the subject of several spectroscopic investigations.
Only a few reports of kinetic measurements of CCO reactions
have appeared, however.8,9 For the CCO + NO reaction, there
is one previous direct measurement; Donnelly et al. used 266
nm laser photolysis of C3O2 and laser-induced fluorescence
(
product channels at moderate temperatures and are not consid-
ered further in this study. The most important secondary reaction
in the CCO + NO system is that of NCO produced in reaction
1
2
-7
1
b with NO:
NCO + NO f CO + N O
(3a)
(3b)
2
f N + CO
2
2
-
11
3
-1
detection to obtain k ) (4.33 ( 0.12) × 10 cm molecule
-
1
9
s
at 298 K. An older relative rate measurement has also been
1
3-18
The kinetics of this reaction have been extensively studied.
10
reported. No information on product channels is available, and
no reports of the CCO + NO2 reaction have appeared.
In this study, we report infrared diode laser measurements
of both the total rate constants (over the temperature range 298-
-
11
3
-1 -1 13-17
At 298 K, k3 ) 3.2 × 10
cm molecule s ,
and the
18
branching ratios are φ3a ) 0.44 ( 0.07 and φ3b ) 0.56 ( 0.07.
CCO radicals were formed in our experiments using pho-
tolysis of carbon suboxide (C3O2). The photodissociation
dynamics of C3O2 has been previously studied by Anderson
5
+
73 K) and the product branching ratios (at 298 K) of the CCO
NO reaction. Ab initio calculations of the potential energy
1
9
and Rosenfeld. They suggest that at 193 nm, the primary
surface of this reaction are also reported, and are useful in the
interpretation of isotopic labeling experiments used to determine
the branching ratio. In addition, we report rate constant
measurements of the CCO + NO2 reaction.
1
photoproducts are ground-state CO and excited-state CCO (a ∆);
3
-
however, at 248 nm, mostly ground-state CCO (X Σ ) is
formed. Most of the experiments reported here used 266 nm
photolysis in order to minimize the formation of excited-state
CCO molecules.
The title reactions may produce several products. For CCO
+
NO, only two are exothermic:
Experimental Section
CCO + NO f CO + CN ∆H298 ) -335 kJ/mol (1a)
2
The time-resolved infrared diode laser technique has been
f CO + NCO ∆H298 ) -328 kJ/mol (1b)
18,20
described previously.
Continuous, high-resolution (0.0003
-
1
f C + NO2 ∆H298 ) 494 kJ/mol
(1c)
(1d)
(1e)
cm ) infrared radiation from a lead-salt diode laser (Laser
Photonics) was made collinear with 266-nm radiation from an
Nd:YAG laser (Lumonics) by means of a dichroic mirror. The
laser beams were then copropagated down a 1.43-m Pyrex
absorption cell. The infrared light then passed through a 0.25-m
monochromator and was focused onto an InSb detector (Cincin-
nati Electronics, ∼1 µs response). Transient signals were
collected and averaged on a digital oscilloscope and stored on
2
f CNC + O2 ∆H298 ) 179 kJ/mol
f CCN + O2 ∆H298 ) 182 kJ/mol
CCO + NO f CN + CO + O2 ∆H298 ) 4.9 kJ/mol (2a)
2
f CO + NCO ∆H ) -585.5 kJ/mol (2b)
2
298
2
1
f NCCO + O2
(2c)
a computer. The HITRAN database and published spectral
2
data for CCO were used as an aid in calibrating laser
wavelengths and identifying transitions.
Carbon suboxide was prepared just prior to each experiment
by reaction of phosphorus pentoxide (P2O5) (Aldrich) with
The thermochemical information for reactions (1a-d) and (2a-
*
Corresponding author. E-mail: john.hershberger@ndsu.nodak.edu.
1
0.1021/jp0304125 CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/04/2003

Kinetics of the CCO + NO and CCO + NO2 Reactions
J. Phys. Chem. A, Vol. 108, No. 1, 2004 75
Figure 1. Transient infrared absorption signal of CCO at 1984.326
cm , produced by 266 nm photolysis of C O . Reaction conditions:
3 2
Figure 2. Pseudo-first-order decay rate constant of CCO radical as a
function of NO pressure. Reaction conditions: PC3O2 ) 0.500 Torr, PNO
) variable, P ) 1.00 Torr, P ) 1.00 Torr. Crosses: 298 K.
-
1
P
P
C3O ) 0.500 Torr, PNO ) 0.0 Torr (trace A), 0.060 Torr (trace B),
2
SF6
Xe
SF6 ) 1.00 Torr, PXe ) 1.00 Torr, T ) 298 K.
Diamonds: 378 K. Triangles: 573 K.
malonic acid (Aldrich) as described in the literature.22 The
resulting gas was then passed slowly through a U-tube filled
with Ca(OH)2 several times, then frozen at 77 K into a 500 mL
Pyrex storage bulb. The resulting gas mixture is usually at least
found rise times substantially slower than those in Figure 1. As
a result, most of the experiments reported here used 266 nm.
As shown in Figure 1, addition of NO significantly increased
the rate of decay of the transient signals, indicating reactive
removal of CCO by reactions (1) or (2). In principle, a fit of
the decay portion of the signal to an exponential function would
yield the pseudo-first-order rate constant k′. This approach
assumes, however, that the rise rates in the signal are very fast
compared to the decay rates. Some of our signals, especially at
high [NO], have decay rates that approach the rise rates. As a
result, the transient signals were fit to a function consisting of
the sum of a rising exponential and a decaying exponential,
using a nonlinear Marquardt algorithm. This procedure yields
decay rates that are typically ∼10% greater than those obtained
from a single decaying exponential fit. The decay rate is then
interpreted as the pseudo-first-order rate constant k′. As per
standard pseudo-first-order kinetics, a plot of k′ as a function
of [NO] yields a straight line given by
60% CO2 due to competing thermal decarboxylation reactions.
Other reagents were obtained from Matheson and purified by
several freeze-pump-thaw cycles. Typical reaction conditions
were 0.1-0.5 Torr C3O2 sample, 0.050-0.80 Torr NO or NO2,
1
.0 Torr SF6, and 1.0 Torr Xe. Upon introduction into the
reaction cell, gases were allowed to stand ∼5 min in order to
ensure complete mixing.
The following transitions were probed:
CCO R(18) line (ν1 vibrational band) at 1984.326 cm
-
1
-
1
CN V ) 0, R(12) line at 2085.043 cm
-
1
N2O (000) P(14) line at 2211.398 cm
1
6
-1
-1
C O V ) 0, P(10) line at 2103.270 cm
C O V ) 0, R(2) line at 2102.9117 cm
O C O (000) R(14) line at 2323.798 cm
O C O (000) P(11) line at 2323.694 cm
1
8
1
8
6
12 18
-1
1
12 18
-1
k′ ) k [NO] + k
1
d
Results and Discussion
Total Rate Constant Measurements. Figure 1 shows a
where kd is the decay rate in the absence of NO (due to reaction
of CCO with itself, the precursor, or diffusion of CCO radicals
-
1
out of the probed reaction zone). k is the desired bimolecular
typical transient signal for CCO detection at 1984.326 cm .
As shown, the signals display a rise time of ∼35 µs followed
by a slow decay. This suggests that most of the nascent CCO
is produced in excited vibrational and/or electronic states, and
the rise time is indicative of the rate of relaxation into the probed
ground state. The buffer gases included in the reaction mixture
were necessary to ensure that this relaxation occurred on a time
scale relatively short compared to the subsequent reaction. SF6
is an efficient collision partner for the vibrational relaxation of
many triatomic molecules,23,24 and xenon is expected to promote
the spin-forbidden process:
1
rate constant, obtained from the slope of the k′ vs [NO] plot.
Figure 2 shows such a plot at several reaction temperatures for
the CCO + NO reaction. Figure 3 shows a similar treatment
for the CCO + NO2 reaction. Figure 4 shows Arrhenius plots
for the two reactions. Both reactions have nearly temperature-
independent rate constants over the range 298-573 K. The data
were fit to the following Arrhenius expressions:
k1(CCO + NO) )
-
10
(
1.66 ( 0.5) × 10 exp[(-337 ( 100)/T]
1
3 -
k (CCO + NO ) )
CCO (a ∆) + Xe f CCO (X Σ ) + Xe
2
2
-
11
(
8.51 ( 1.5) × 10 exp[(-63 ( 80)/T]
Experiments performed without Xe buffer gas yielded signals
similar in shape to those of Figure 1, but substantially smaller
in magnitude. This suggests that some excited electronic state
CCO is produced even at our photolysis wavelength of 266 nm,
which is somewhat contradictory to the suggestion of Anderson
and Rosenfeld that primarily ground electronic state CCO is
produced at the nearby 248 nm wavelength.19 We did perform
a few experiments using 193 nm photolysis wavelengths, but
-
11
At 298 K, the rate constants are k1 ) (5.36 ( 0.5) × 10 and
k2 ) (6.89 ( 0.5) × 10 cm molecule s . One previous
direct study reported a somewhat lower value of k1 ) (4.33 (
0.12) × 10 cm molecule
-
11
3
-1 -1
-11
3
-1 -1
9
s
at 298 K, using LIF detection
of CCO. Our work represents the first reported value of k2.
Product Channel Measurements. All of the reactants and
products in channels (1a) and (1b) may in principle be detected

76
J. Phys. Chem. A, Vol. 108, No. 1, 2004
Thweatt et al.
1
8
12 18
Figure 5. Transient infrared absorption signals of O C O (000)
Figure 3. Pseudo-first-order decay rate constant of CCO radical as a
function of NO pressure. Reaction conditions: PC3O2 ) 0.500 Torr,
-
1
18 12 16
R(14) at 2323.798 cm (trace A) and O C O (000) P(11) at
2
-
1
2
323.694 cm (trace B). Reaction conditions: PC3O ) 0.100 Torr,
P
NO2 ) variable, PSF6 ) 1.00 Torr, PXe ) 1.00 Torr. Crosses: 298 K.
2
P
NO ) 0.200 Torr, PSF6 ) 1.00 Torr, PXe ) 1.00 Torr, T ) 298 K.
Diamonds: 378 K. Triangles: 573 K.
(1a) is a minor product channel. CO2 detection by infrared
spectroscopy is much more sensitive than CN detection,
however, so the experiments described below give a much better
estimate of the importance of channel (1a).
The following approach was found to provide the most
reliable estimate of the product branching ratio of reaction (1).
18
CCO was allowed to react with isotopically labeled N O:
1
8
16 12 18
CCO + N O f CN + O C O
(1a)
(1b)
1
6
18
f C O + NC O
The secondary reaction (3) then produces doubly labeled CO2:
1
8
18
18
18
NC O + N O f C O + N
O
(3a)
(3b)
2
1
8
12 18
Figure 4. Arrhenius plots for the CCO + NO (crosses) and CCO +
NO (diamonds) reactions.
2
f N + O C O
2
Because we know the branching ratio of reaction (3), i.e., φ3b
0.56, one can obtain the yield of channel (1b) by measuring
by infrared absorption spectroscopy. The nontransient species
CO and CO2 are the easiest to quantify. It may therefore appear
that φ1a and φ1b may be readily obtained by quantifying the
yields of these products. Unfortunately, several sources of
background signals complicate this approach. First, CO2 is
present as a significant impurity in the C3O2 samples. Attempts
to remove CO2 by flowing the sample through an Ascarite trap
were unsuccessful, apparently resulting in decomposition of the
C3O2. We note that even a few percent impurity will signifi-
)
18 12 18
18
[
O C O], if [N O] is sufficiently high to ensure that every
18
18
NC O radical reacts with N O:
1
8
18 12 18
[
NC O] ) [ O C O]/0.56
1
6
12 18
Measurement of [ O C O] provides a direct measurement of
1
6
12 18
the yield of channel (1a). Comparison of [ O C O] and
1
8
[NC O] then yields the branching ratios φ1a and φ1b. This
procedure thus separately measures the CO2 yield produced by
the direct reaction (1a) from that produced by the secondary
chemistry, channel (3b), without undue interference from the
unlabeled CO2 impurity present in the C3O2 sample.
The procedure just described relies on the crucial assumption
that in channel (1b), the 18-labeled oxygen from the NO reactant
ends up on the NCO and not the CO product. Ab initio
computations, described below, strongly suggest that this
assumption is valid.
cantly interfere with any attempt to detect 1 O C O reaction
products. Isotopic substitution of reagents is therefore an
absolute necessity in order to quantify the CO2 yield. Detection
of CO is also problematic, because it is formed in the photolysis
6
12 16
(
presumably with unity quantum yield). As a result, a large
transient signal for CO was observed upon the photolysis of a
C3O2/buffer gas mixture, without any NO reagent. If NO is
included in the reaction mixture, additional CO formation routes
(1b) and the secondary reaction (3a) are available. We therefore
expected the CO yield to be substantially increased when NO
was included. In fact, no such increase was observed; the CO
yield was found to be essentially unchanged over the range
Figure 5 shows transient signals for the singly and doubly
1
5
18
labeled isotopes of CO2 upon photolysis of C3O2/ N O/SF6/
15
18
Xe mixtures. ( N O was used because it was readily available;
the nitrogen labeling is irrelevant in this experiment). The
0
-0.5 Torr NO. Our explanation of this is that even in the
1
8
12 18
16 12 18
absence of nitric oxide reagent, CCO can undergo reactions,
either with itself or with C3O2, ultimately resulting in the
formation of CO. It is therefore difficult to determine how much
of the observed CO originated from channel (1b).
O C O trace is an average of 10 shots, and the O C O is
16
12 18
an average of 40 shots. The signal for [ O C O] is somewhat
noisy because the CO2 impurity in the C3O2 sample has a small
but nonzero natural abundance of this isotope, resulting in a
significant but manageable level of static background. The peak
amplitudes of the transient signals were converted into absolute
Attempts were made to detect CN radicals produced in
channel (1a). No transient signals were found, suggesting that

Kinetics of the CCO + NO and CCO + NO2 Reactions
J. Phys. Chem. A, Vol. 108, No. 1, 2004 77
Figure 6. 18O C O (diamonds) and O C O (circles) product yields
12
18
16 12 18
1
8
as a function of N O pressure. Reaction conditions: PC3O2 ) 0.100
Torr, PNO ) variable, PSF6 ) 1.00 Torr, PXe ) 1.00 Torr, T ) 298 K.
Figure 8. Another reaction path producing CO + NCO. Calculation
at same level of theory as Figure 7.
Figure 7. A portion of the QCISD(T)/6-311G(2df,2pd)//MP2/6-31G*
potential energy surface of CCO + NO: a reaction path producing
CO + NCO.
number densities using formulas previously described. One
modification to that procedure was necessary: since HITRAN
assumes natural isotopic abundances, linestrengths from the
database were modified by dividing by the natural abundance
-
6
of the detected isotope (0.0039 for singly labeled, 4 × 10 for
doubly labeled). Figure 6 shows the resulting product yields as
a function of NO pressures. The product yields and branching
ratios were found to be essentially independent of NO pressure
over the range 0.05-0.8 Torr. The average of 18 experiments
over this pressure range gives the following results: φ1b/φ1a )
2
Figure 9. A reaction path producing CN + CO . Calculation at same
level of theory as Figure 7. Also shown is part of a high energy pathway
involving an NOCCO structure (see text).
become more important. The observation in Figure 6 of constant
product yields over the range 0.05-0.8 Torr of NO suggests,
however, that this competition between ground- and excited-
state reactions does not significantly affect our results, probably
because the product yields of excited-state reaction with NO
are similar to those of ground-state reactions.
Donnelly et al. suggested that a dark reaction between C3O2
and NO may have occurred in their experiments, as evidenced
by a slow pressure rise upon mixing of the reagents, as well as
their observation of otherwise unexplained fluorescence signals
upon 266-nm excitation of C3O2/NO. We did not observe any
pressure increases in our experiments, but a slow secondary
reaction between C3O2 and NO could result in some depletion
of reagents. Depletion of C3O2 would merely reduce the initial
CCO radical concentration, without affecting the pseudo-first-
order kinetics or branching ratio measurements. Loss of NO,
6
.69 ( 2.0. If we assume that only channels (1a) and (1b)
contribute, we obtain the absolute branching ratios φ1a ) 0.13
(
0.05 and φ1b ) 0.87 ( 0.05, where the error bars represent
one standard deviation.
In principle, the formation of vibrationally and/or electroni-
cally excited radicals in the photolysis step can affect product
yield determinations. As Figure 1 shows, at low NO pressures
(
∼0.05-0.1 Torr), the rate of reaction with NO (the decay rate)
is somewhat slower than the rate of relaxation of CCO into the
ground vibrational and electronic state. Under these conditions,
the CCO + NO reaction therefore predominates over excited-
state CCO* + NO reaction events. At higher NO pressures (not
shown in Figure 1), the rate of reaction with NO is presumably
faster than the relaxation rate (which is primarily governed by
collisions with buffer gas molecules), so CCO* + NO reactions

78
J. Phys. Chem. A, Vol. 108, No. 1, 2004
Thweatt et al.
TABLE 1: Ab Initio Calculations of Reactants and
Productsa
on the doublet reaction surface. Unless otherwise noted, the
25
geometries were calculated using the MP2 method with
2
6
species
S+1
symmetry
CCO
3
NO
2
CO2
1
CO
1
CN
2
NCO
2
Pople’s 6-31G* basis set (MP2/6-31G*). Stationary points
minima and transition states) were confirmed with vibrational
(
2
analysis. Zero-point vibrational energy corrections were calcu-
C
∞V
C
∞V
D
∞h
C
∞V
C
∞V
C
∞V
27
lated at the Hartree-Fock level of theory with the same basis
2
8
R(C
R(C
R(C
1
2
2
-C
2
)
1.379
1.173
set, then scaled by a factor of 0.8978. Absolute electronic
energies were calculated at the MP2/6-31G* geometry using
quadratic configuration interaction with singles, doubles, and
-O
-O
1
)
)
1.180
1.180
2
1.151
R(N-O
1
)
1.143
2
9
perturbative triples excitations with Pople’s 6-311G(2df,2pd)
R(C
R(C
1
-N)
-O
1.136
1.255
1.167
3
0
1
1
)
basis set (QCISD(T)/6-311G(2df,2pd)). In two cases (I5 and
TS5), due to massive MP2 spin-contamination problems,
a
Bond lengths in Angstroms, angles in degrees.
transition-state geometries were refined at the B3LYP/6-31G*
TABLE 2: Ab Initio Calculations of Intermediatesa
31,32
and B3LYP/6-311G* levels of theory,
respectively. All ab
species
S+1
I1
2
I2
2
I3
2
I4
2
I5
2
IP1
2
initio quantum chemical calculations were performed on IBM
RS6000 3CT machines using Gaussian 94, version E.33
2
Figures 7, 8, and 9 show important stationary points along
the calculated minimum energy pathways on the ground-state
doublet PES. As shown, N f C attack can produce a bent
ONCCO complex (I1) which lies 234 kJ/mol below the
reactants. Several possible pathways from I1 exist. One route
symmetry
Cs
C1
Cs
Cs
C2V
Cs
R(O1-N)
1.195 1.442 1.512 1.316
1.273 1.375 1.216
1.352 1.294 1.592 1.367 1.460
1.433
R(N-C1)
1.161 1.360
R(C1-C2)
R(C2-O2)
1.164 1.146 1.179 1.171 1.254
R(C2-O1)
1.254
1.351
(Figure 7) involves formation of an (NOC)CO (three-membered
R(O1-C1)
1.384
63.5
A(O1-N-C1)
A(N-C1-C2)
A(C1-C2-O2)
A(N-O1-C1)
A(O1-C1-C2)
A(O1-C2-O2)
D(O1-N-C1-C2)
D(N-C1-C2-O2)
D(N-O1-C1-C2)
D(O1-C1-C2-O2)
138.6 62.0
112.4 102.8 93.4
168.4 130.0 142.0 160.2 122.4
92.9
ring) structure (I2), followed by C-C bond fission to produce
CO and a cyclic (ONC) structure (IP1). Opening of the three-
membered ring can then produce the NCO + CO product
channel. Alternatively, as shown in Figure 8, the bent ONCCO
complex (I1) could dissociate directly to CNO + CO, and the
CNO then may have enough energy to isomerize, via structure
IP1, to NCO. Channel (1a) is accessible by a route shown in
Figure 9, involving formation of a four-membered NOCCO
structure (I3), followed by a ring-opening N-O bond fission
to structure I5, and subsequent C-C bond fission to CN + CO2.
These pathways are all energetically accessible from the CCO
180.0 57.5
121.0
121.7 122.4
115.2
180.0 134.9 0.0
0.0 163.4 180.0
0.0
0.0
180.0
a
Bond lengths in Angstroms, angles in degrees.
however, would cause a systematic error in the total rate constant
1
8
+
NO starting energy, and if N O is used, the labeled oxygen
determinations. To test for this possibility, we tuned the diode
laser to weak NO transitions near 1900 cm-1 to monitor [NO]
would end up on the NCO product in (1b), rather than CO.
This is therefore consistent with the assumption used to interpret
our isotopic labeling experiments. The most obvious reaction
path that would violate this assumption is also shown in Figure
9, and would involve formation of an NOCCO intermediate (I4),
which could then rearrange via a four-membered ring (not
shown), followed by N-O and C-C bond fission to form NCO
over a ∼1 h time scale upon mixing of reagents. Only a very
slight loss of NO was observed, and was insignificant over the
5-10 min time scale of our measurements. This suggests that
dark reactions, if any, occur too slowly to affect our data.
Ab Initio Calculations. To better understand the reaction
mechanism, and to help interpret the branching ratio experi-
ments, ab initio quantum chemical calculations were performed
1
8
+ CO products. If this pathway were active, CCO + N O
TABLE 3: Ab Initio Calculations of Transition Statesa
species
S+1
symmetry
R(O -N)
R(N-C
TS1
2
TS2
2
TS3
2
TS4
2
TS5
2
TS6
2
PTS1
2
PTS2
2
2
C
1
C
1
C
S
C
S
C
S
C
S
C
S
C
S
1
1.330
1.352
1.390
1.156
1.425
1.374
1.917
1.146
1.366
1.295
1.458
1.134
1.513
1.216
1.593
1.179
1.180
2.253
1.438
1.316
1.438
1
)
)
1.162
1.926
1.211
1.163
R(C
R(C
R(C
R(O
A(O
1
2
2
-C
2
1.385
1.172
-O
-O
2
)
)
)
1.151
1.151
1
1
1
-C
1.836
1.402
36.9
2.250
109.5
1
-N-C
-C
-O
-C
-C
-O
-N-C
-C
-C
-C
1
)
78.2
134.2
146.3
62.0
102.8
130.3
98.8
105.3
151.4
92.9
93.4
142.0
A(N-C
A(C -C
A(N-O
1
2
)
171.3
88.1
1
2
2
)
172.6
122.2
96.8
1
1
)
2
)
A(O
A(O
D(O
1
1
1
-C
-C
1
2
2
)
157.9
1
-C
2
2
2
)
)
)
2
89.7
171.0
134.9
163.4
0.0
180.0
0.0
180.0
D(N-C
D(N-O
D(O -C
1
2
-O
1
1
-C
180.0
180.0
180.0
180.0
1
1
2
-O
)
a
Bond lengths in Angstroms, angles in degrees.

Kinetics of the CCO + NO and CCO + NO2 Reactions
J. Phys. Chem. A, Vol. 108, No. 1, 2004 79
18
would produce NCO + C O, in contrast to our assumptions.
The most important result from the ab initio calculations is that
the NOCCO structure is endoergic by 42 kJ/mol relative to the
CCO + NO reactants. Furthermore, the most direct path to this
structure, via TS6, involves a 88 kJ/mol barrier. Although other
pathways (not calculated) to I4 may exist (for example, opening
of the NOCCO three-membered ring (I2) via N-C bond fission),
the NOCCO structure itself is sufficiently high in energy to
exclude these pathways as major contributors to the reaction
mechanism.
(3) Moazzen-Ahmadi, N.; Boere, R. T. J. Chem. Phys. 1999, 110, 955.
(
(
4) Moazzen-Ahmadi, N.; Boere, R. T. J. Chem. Phys. 1998, 108, 6588.
5) Ohashi, N.; Kiryu, R.; Okino, S.; Fujitake, M. J. Mol. Spectrosc.
1
993, 157, 50.
(6) Fujitake, M.; Kiryu, R.; Ohashi, N. J. Mol. Spectrosc. 1992, 154,
169.
(7) Becker, K. H.; Horie, O.; Schmidt, V. H.; Wiesen, P. Chem. Phys.
Lett. 1982, 90, 64.
(8) Bauer, W.; Becker, K. H.; Meuser R. Ber. Bunsen-Ges. Phys. Chem.
1
985, 89, 340.
(9) Donnelly, V. M.; Pitts, W. M.; McDonald, J. R. Chem. Phys. 1980,
49, 289.
10) Williamson, D. G.; Bayes, K. D. J. Am. Chem. Soc. 1967, 89, 3390.
11) Chase, M. W., Jr. J. Phys. Chem. Ref. Data, Monograph No. 9,
NIST-JANAF Thermochemical Tables, 4th ed.; 1998.
12) Cyr, D. R.; Continetti, R. E.; Metz, R. B.; Osborn, D. L.; Neumark,
D. M. J. Chem. Phys. 1992, 97, 4937.
(13) Perry, R. A. J. Chem. Phys. 1985, 82, 5485.
(14) Hancock, G.; McKendrick, K. G. Chem. Phys. Lett. 1986, 127, 125.
15) Atakan, B.; Wolfrum, J. Chem. Phys. Lett. 1991, 178, 157.
16) Mertens, J. D.; Dean, A. J.; Hanson, R. K.; Bowman, C. T. Symp.
Int.) Combust. Proc. 1992, 24, 701.
(17) Juang, D. Y.; Lee, J.-S.; Wang, N. S. Int. J. Chem. Kinet. 1995,
7, 1111.
18) Cooper, W. F.; Park, J.; Hershberger, J. F. J. Phys. Chem. 1993,
7, 3283.
19) Anderson, D. J.; Rosenfeld, R. N. J. Chem. Phys. 1991, 94, 7857.
20) Quandt, R. W.; Hershberger, J. F. J. Phys. Chem. 1996, 100, 9407.
(21) Rothman, L. S.; Gamache, R. R.; Tipping, R. H.; Rinsland, C. P.;
Smith, M. A. H.; Benner, D. C.; Devi, V. M.; Flaud, J. M.; Camy-Peyret,
C.; Perrin, A.; Goldman, A.; Massie, S. T.; Brown, L. R.; Toth, R. A. J.
Quant. Spectrosc. Radiat. Transfer 1992, 48, 469.
(
(
One limitation of both the calculations and the experiments
reported above is that we cannot rule out the possibility that a
major product is the high energy isomer CNO rather than NCO.
We have no direct method of detecting CNO, however the
following experiment was performed. We measured the N2O
(
1
8
12 18
yield using natural abundance samples, and the O C O yield
(
(
18
using N O labeling. Since these products are formed only in
reactions (3a) and (3b), the ratio of these product yields should
be the same as previously measured branching ratios of reaction
(
2
9
(
3). We obtain φ3a ) 0.49 ( 0.02 and φ3b ) 0.51 ( 0.02, in
reasonable agreement with our previous measurements of φ3a
0.44 and φ3b ) 0.56.18 This strongly suggests one of the
(
)
(
(
following: either the (NCO) product formed in the title reaction
is NCO and not CNO, or any CNO quickly rearranges to NCO,
or that the CNO + NO reaction (for which no literature data
exist) has branching ratios similar to NCO + NO.
(
22) Long, D. A.; Murfin, F. S.; Williams, R. L. Proc. R. Soc. London
Conclusions
Ser. A 1954, 223, 251.
(
(
(
23) Fakhr, A.; Bates, R. D., Jr. Chem. Phys. Lett. 1980, 71, 381.
24) Stephenson, J. C.; Moore, C. B. J. Chem. Phys. 1970, 52, 2333.
25) Krishnan, R.; Pople, J. A. Int. J. Quantum Chem. 1978, 14, 91.
The kinetics of the CCO + NO and CCO + NO2 reactions
were investigated. Both reactions are fast, with nearly temper-
ature-independent rate constants. Product yield experiments
using isotopically labeled NO combined with ab initio calcula-
tions demonstrate that NCO + CO is the major product channel
of the CCO + NO reaction at 298 K, with a branching ratio of
(26) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 65,
257.
27) Fock, V. Z. Physik 1930, 61, 126.
28) Curtis, L. A.; Raghavachari, K.; Trucks, G. W.; Pople, J. A., J.
2
(
(
Chem. Phys. 1991, 94, 7221.
(29) Pople, J. A.; Head-Gordon, M.; Raghavachari, K. J. Chem. Phys.
1987, 87, 5968.
0
.87 ( 0.05. CN + CO2 is a minor channel, with a branching
ratio of 0.13 ( 0.05.
(30) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys.
1
980, 72, 6.
Acknowledgment. This work was supported by the Division
of Chemical Sciences, Office of Basic Energy Sciences of the
Department of Energy, Grant DE-FG03-96ER14645.
(
(
(
31) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
32) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
33) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T. A.; Petersson,
G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakreze-
wski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzales, C.; Pople, J. A. Gaussian 94; Gaussian, Inc.:
Pittsburgh, PA, 1995.
References and Notes
(
1) Miller, J. A.; Bowman, C. T. Prog. Energy Combust. Sci. 1989,
5, 287.
2) Yamada, C.; Kanamori, H.; Horiguchi, H.; Tsuchiya, S.; Hirota,
E. J. Chem. Phys. 1986, 84, 2573.
1
(