The reaction of C6H5 with CO: Kinetic measurement and theoretical correlation with the reverse process

Article abstract of DOI:10.1021/jp993245a

Phenyl radicals play a pivotal role in the combustion of gasoline in which small aromatic hydrocarbons (BTEX) are added as antiknock agents (≤ 30%). The C₆H₅ radical is involved in the formation of PAH, which are precursors of soot. The kinetics of the C₆H₅ reactions with CO was studied by the cavity ring-down spectrometric technique at 295-500 K and 12-120 torr with argon as the carrier gas. To correlate the kinetic data with those of Solly and Benson, quantum chemical calculations were performed to provide the molecular and transition-state parameters for a more reliable prediction of the temperature and pressure effects on the rate constants for forward and reverse directions by using Rice-Ramsperger-Kassel-Marcus theory. Combination of the two sets of kinetic data yielded the heat of reaction, C₆H₅ + CO = C₆H₅CO (1), ΔH₁°= -24.6 kcal/mole by the third-law method. This result, combined with the known heats of formation of the reactants, gave rise to the heat of formation of the benzoyl radical, 32.5 ± 1.5 kcal/mole at 0 K.

Full text of DOI:10.1021/jp993245a

J. Phys. Chem. A 2000, 104, 1233-1239
1233
The Reaction of C6H5 with CO: Kinetic Measurement and Theoretical Correlation with the
Reverse Process
†
Gi-Jung Nam, Wensheng Xia, J. Park, and M. C. Lin*
Department of Chemistry, Emory UniVersity, Atlanta, Georgia 30322
ReceiVed: September 13, 1999; In Final Form: December 1, 1999
6 5
The kinetics of the C H reaction with CO has been studied by the cavity ring-down spectrometric technique
in the temperature range 295-500 K at 12-120 Torr pressure with Ar as the carrier gas. The reaction occurred
near the high-pressure limit under the conditions used. A weighted least-squares analysis of all data gives the
1
1.93(0.14
3
-1 -1
rate constant for the association reaction, k
1
) 10
exp[-(1507 ( 109)/T] cm mole s ), where the
errors represent two standard deviations. Our result can correlated satisfactorily with the kinetic data reported
by Solly and Benson for the reverse process (J. Am. Chem. Soc. 1971, 93, 2171) with the Rice-Ramsperger-
Kassel-Marcus (RRKM) theory using the transition-state parameters computed quantum mechanically by
the MP2 method with the 6-31G(d,p) basis set. Combination of the forward and reverse reaction data gives
∞
-1
14
-1
∆
3
H° ) -24.6 ( 0.8 kcal/mol at 0 K and k ) 5.3 × 10 exp(-14 600/T) s for the temperature range
1
00-670 K. The heat of reaction, combining with the known heats of formation of the reactants, leads to
∆fH° (C CO) ) 32.5 ( 1.5 kcal/mol.
6 5
H
0
Introduction
molecular and transition-state parameters for a more reliable
prediction of the temperature (T) and pressure (P) effects on
the rate constants for both forward and reverse directions by
means of the RRKM theory. The results of this investigation
are reported here.
Phenyl radicals play a pivotal role in the combustion of
gasoline in which small aromatic hydrocarbons (benzene,
1
3
toluene, and xylenes) are added as antiknock agents (up to
1
3
0%). The C6H5 radical is also believed to be involved in the
formation of polycyclic aromatic hydrocarbons (PAHs), which
are precursors of soot.2
-4
Experimental Procedure
Phenyl radicals and carbon monoxide coexist under sooting
conditions; their interaction may affect the concentration of C6H5
and thus the chemistry of soot formation. The kinetics of the
C6H5 + CO f C6H5CO reaction 1 is unknown, although the
rate constant for the reverse process, the decomposition of the
The experimental apparatus and kinetic data acquisition by
CRDS have been discussed in detail in our previous publica-
6
-12
tions.
Two pulsed lasers were used for the pump and probe
processes. A Lambda Physik LPX 105E excimer laser was used
to photodissociate C6H5NO (nitrosobenzene) at 248 nm. A
tunable, pulsed dye laser pumped by XeCl excimer laser
5
benzoyl radical, was reported by Solly and Benson more than
30 years ago. In their study with the I2-catalyzed decomposition
(Lambda Physik FL 3002) was used for probing the C6H5
of benzaldehyde in a narrow temperature range of 614-667 K,
∞
14
directly at 504.8 nm where a distinct absorption peak was known
the first-order (high-pressure) rate constant, k ) 4 × 10
-
1
6
-
1
to exist. Nitrosobenzene was placed on a sealed, fritted glass
exp(-14 800/T) s , was evaluated by extrapolating the rate
coefficients obtained in the falloff region using assumed
transition-state structure and vibrational frequencies by both
Rice-Ramsperger-Kassel (RRK) and Rice-Ramsperger-
disk and carried into the reactor via a mixing tube with Ar as
a carrier gas. The decay signal was measured with a Hamamatsu
photomultiplier and acquired and averaged with a multichannel
digital oscilloscope (LeCroy 9310M). The averaged signal was
sent to a microcomputer for storage and further analysis. A pulse
delay generator (SR DG535), interfaced with computer, was
used to control the delay time between the two laser firings
and the triggering of the oscilloscope. The lasers were typically
operated at 4 Hz. The detailed experimental conditions, such
as flow rates, pressure, and temperatures, used in the present
study are presented in Table 1. The temperature of the cavity
was controlled by resistive heating and measured using a J-type
thermocouple placed a few millimeters below the central axis
of the cavity. The amount of CO and the carrier gas were
precisely measured with calibrated MKS mass flowmeters.
5
Kassel-Marcus (RRKM) theories. Their result was consistent
with the heat of the dissociation reaction at 298 K, ∆H° )
-
1
2
7.5 kcal/mol, with a 2.3 kcal/mol C6H5 + CO addition barrier.
To our knowledge, no other kinetic data exist for the reaction
in either direction since this early study.
In this investigation, we apply the technique of cavity ring-
down spectrocopy (CRDS), which has been developed for
6-12
kinetic applications in our laboratory since 1992
to measure
the rate constant for the C6H5 + CO addition reaction over the
temperature range 295-500 K at 12-120 Torr pressure. To
5
correlate our kinetic data with those of Solly and Benson, we
also performed quantum chemical calculations to provide the
C6H5NO (Aldrich, 97%) was recrystallized using ethanol as
solvent and vacuum-dried. The CO reactant was purified by
passing through the silica gel trap at 180 K [to remove impurities
such as Fe(CO)5] and introduced into the cavity without carrier
*
Corresponding author. E-mail: chemmcl@emory.edu.
Permanent address: State Key Laboratory for Physical Chemistry of
†
Solid Surfaces, Institute of Physical Chemistry and Department of Chem-
istry, Xiamen University, Xiamen, 361005, P. R. of China.
1
0.1021/jp993245a CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/21/2000

1234 J. Phys. Chem. A, Vol. 104, No. 6, 2000
Nam et al.
TABLE 1: Measured Rate Constants and Heat of Reactionb
a
for the C
6
H
5
6 5
+ CO f C H CO Reaction
-
10
T (K)
P (Torr)
P
co (Torr)
k
1
× 10
∆H1°
c
c
c
c
c
c
c
d
d
d
d
d
d
295
319
336
347
347
347
347
369
397
419
442
467
500
40
40
40
12
40
80
120
40
40
40
40
40
40
0 - 11.10 0.51 ( 0.08
0 - 10.30 0.80 ( 0.12
-24.65 ( 0.09
-24.67 ( 0.09
-24.61 ( 0.07
-24.65 ( 0.12
-24.63 ( 0.12
-24.59 ( 0.09
-24.60 ( 0.04
-24.56 ( 0.27
-24.44 ( 0.16
-24.45 ( 0.13
-24.54 ( 0.05
-24.66 ( 0.15
-24.60 ( 0.39
-24.59 ( 0.14
0 - 9.75
0 - 4.27
0 - 8.97
0.96 ( 0.10
1.14 ( 0.22
1.16 ( 0.22
0 - 11.80 1.13 ( 0.16
0 - 14.90 1.11 ( 0.07
0 - 7.40
0 - 8.26
0 - 6.00
0 - 5.68
0 - 5.85
0 - 2.95
1.41 ( 0.64
1.70 ( 0.38
2.16 ( 0.37
2.95 ( 0.17
4.02 ( 0.70
4.56 ( 2.20
average
Figure 1. Typical pseudo-first-order decay plots for the C
6 5
H +
a
Rate constants are given in units of cm mole s-1 and their
3
-1
CO reaction under different, excess reactant concentrations in units
b
associated errors represent one standard deviations. Evaluated by the
third-law method (see below) given in units of kcal/mol. Analyzed
by eq 1. Analyzed by eq 6.
3
-8
of mol/cm at 336 K. O, [CO] ) 1.66 × 10 ; 0, [CO] ) 9.75 ×
c
-8
1
0
.
d
Equation 2 may not be valid for the C6H5 + CO reaction at
gas. Ar (Specialty Gases, 99.995% UHP grade) was used to
carry the vapor of nitrosobenzene into the cavity without further
purification.
higher temperatures because of the inception of the reverse
reaction (-1):
Data Acquisition. The CRDS method measures the decay
times of injected probing photons in the absence (tc ) and
C H + CO T C H CO (1, -1)
6
5
6
5
0
presence (tc) of absorbing species. Because of the high-quality
In this case, the rate equation should include the reverse term:
dx/dt ) k ([A] - x)[CO] - k x (4)
resonant cavity used, the injected dye laser pulse was typically
3
0
lengthened by a factor of 2-3 × 10 , giving tc = 20 µs in the
absence of an absorbing species, denoted by A. In the presence
of the absorbing species such as C6H5, the photon decay time
1
0
-1
where x ) [A]0 - [A]t, and [A]t represents the concentration
of the benzoyl radical at time t. Integration of eq 4 under excess
CO concentration conditions gives
0
at 504.8 nm (tc) was shortened by about 25% from that of tc
immediately after photodissociation. Because the measured
0
values of tc and tc are much smaller than the chemical decay
-
bt
x ) (a/b) (1 - e
)
(5)
time of the radical, t, which depends on the concentration of
6,7
the molecular reactant, the following kinetic relationship holds
where a ) k1 [CO][A]0 and b ) k1[CO] + k-1. Combining eqs
and 5 leads to the following general expression for the
reversible reaction:
1
0
1
/t ) 1/t + (clꢀ/nL)[A]
(1)
c
c
t
0
-bt
where c is the velocity of light, l is the length of the absorbing
medium, ꢀ is the extinction coefficient, n is the refractive in-
dex of the medium, L is the length of the cavity, and [A]t is
the concentration of the absorbing species at time t. For a sim-
ple association reaction with a deep well (without the prob-
lem of redissociation), A decays exponentially with [A]t )
1/t -1/t ) k d/b + (k [CO]d/b)e
(6)
c
c
-1
1
In eq 6, d ) 1/tc0 - 1/t ; where t is the photon decay time
0
0
c
c
measured immediately after the photolysis of C H NO at t )
6
5
0. Because d is proportional to the initial concentration of C H ,
6
5
it is a constant for each experimental run. Equation 6 will be
used to analyze data obtained at higher temperatures at which
the redissociation of C H CO becomes significant.
-
k′t
[A]0 e
.
Combination with eq 1 leads to6
-12
6
5
0
Results and Discussion
ln(1/t - 1/t ) ) B - k′t
(2)
c
c
A. Kinetics of C6H5 + CO. The rate constants for the C6H5
+ CO reaction were measured in the temperature range between
295 and 500 K at 12-120 Torr pressure with Ar as a carrier
gas. Figure 1 presents typical pseudo-first-order plots for the
decay of C H at 336 K in the presence of the different amounts
where B ) ln {(clꢀ/nL)[A]0}. The validity of eq 2 has been
illustrated previously6 ; the equation will be used later for
analysis of low-temperature data. According to eq 2, the slope
-12
0
of a ln(1/tc - 1/tc ) vs t plot gives the first-order rate coefficient,
6
5
k′, for the decay of A, or the C6H5 radical, in the presence of a
specified molecular reactant concentration, [CO]. A standard
plot of k′ vs [CO] provides the second-order rate constant, k1,
according to the relationship:
of CO at 40 Torr total pressure, as indicated in the caption of
the figure. As shown in the figure, the data exhibit the good
0
linear quality of the ln(1/t - 1/t ) vs t plot as predicted by eq
c
c
1. The slope of the plot, the pseudo-first-order rate coefficient
(
k′) of the reaction at each CO concentration, was obtained by
k′ ) k + k [CO]
(3)
0
1
a standard weighted least-squares analysis. At temperatures
higher than 400 K the first-order decay plots in the present
system become nonlinear with time t as illustrated in the inset
of Figure 2. The deviation can be attributed to the regeneration
of C6H5 by the decomposition of C6H5CO at higher temperatures
because of the low thermal stability of the radical product, as
where k0 is the first-order decay coefficient of the radical in
the absence of the molecular reactant; it is a convoluted decay
rate, consisting of the losses by diffusion, pumping, and
recombination reactions.

Kinetics of C6H5 Reaction with CO
J. Phys. Chem. A, Vol. 104, No. 6, 2000 1235
Figure 4. Arrhenius plot of k
line is the result of a weighted least-squares analysis. Circles, k
evaluated by eqs 1 and 2; squares, k evaluated by eqs 7 and 2 to
minimize the effect of the reverse reaction.
1
at the total pressure 40 Torr. The solid
1
1
1
1.93(0.14
exp[(-1507 ( 109)/T] cm mole s-1 (8)
3
-1
0
Figure 2. Typical 1/t
6
c
- 1/t
c
vs time plots at 446 K according to eq
k1 ) 10
. Curves are least-squares fits. Inset: plots showing deviation from
eq 1 for the same sets of data.
for the temperature range 295-500 K at 40 Torr Ar pressure.
The errors cited above represent two standard deviations.
We have also examined the effect of pressure on the reaction
at 347 K by varying the total pressure from 12 to 120 Torr. As
shown in Figure 5, the effect is negligible under the conditions
studied. At higher temperatures, however, the pressure effect
becomes significant as illustrated by the predicted pressure
dependence presented in the inset of the figure. At the highest
temperature studied, 500 K, k1 is about 70% of the high-pressure
limit. After the correction for the falloff effect with the RRKM
theory to be discussed below, the high-pressure, second-order
rate constant can be described by the equation:
∞
1
12.17(0.18
exp[(-1676 ( 149)/T] cm mole s-1 (9)
3
-1
k ) 10
Figure 3. k′ (open symbols) or b (solid symbols) vs [CO] at different
temperatures. Linear least-squares fit yields the second-order rate
constant k .
1
B. Theoretical Analysis of Kinetic Data. The measured
kinetic data for the C6H5 + CO reaction and those of Solly and
5
Benson for the reverse process have been correlated with the
RRKM theory using the molecular and transition-state param-
eters computed by different quantum chemical methods.
Quantum Calculations. The geometries of the reactants,
transition state, and product have been optimized using two
different methods, MP2 and B3LYP (i.e. Becke’s three-
discussed in the preceding section. These nonlinear data, if
analyzed with eq 1, underpredict the values of k1 as one would
expect. Accordingly, all data acquired near and above 400 K
were evaluated with eq 6, which can be effectively represented
by
14
parameter nonlocal-exchange functional with the nonlocal
1
5
correlation functional of Lee et al.). Vibrational frequencies
have been used for characterization of stationary points, zero-
point energy (ZPE) corrections, and RRKM calculations. In the
RRKM calculations, vibrational frequencies obtained by B3LYP/
0
-bt
1
/t - 1/t ) c + c e
(7)
c
c
1
2
where c1 and c2 are constants for each experimental run and b
k1[CO] + k-1 as defined before.
1
6
6
-311G(d,p) and MP2/6-31G(d,p) were scaled by 1.0 and
)
1
7
0
0.9434, respectively. All the stationary points have been
positively identified for stable species (with the number of
imaginary frequencies NIMAG ) 0) and transition state (with
NIMAG ) 1). All calculations were carried out with the
The nonlinear fitting of (1/tc - 1/tc ) against t according to
eq 7, illustrated in Figure 2 by solid curves, gives values of b
which vary linearly with [CO].
The second-order plots of k′ or b vs [CO] are presented in
Figure 3. The slopes of these plots give the bimolecular rate
constants for the C6H5 + CO reaction. The values of k1 evaluated
for 40 Torr pressure with eqs 1 and 7 are presented in Figure 4
by circles and squares, respectively. As is evident from the
figure, the values of k1 evaluated with the two equations differ
greatly above 400 K because of the redissociation reaction. The
values obtained with the two methods converge at temperatures
where redissociation becomes negligible.
17
Gaussian 94 program.
Both MP2 and B3LYP methods predict similar structures for
the ground-state C6H5CO radical. However, for the transition
state, B3LYP gives a much longer dissociating C-C bond
(2.493 Å vs 2.215 Å by MP2) and thus a much looser TS
structure which predicts a considerably higher A-factor than the
experimentally measured value. Its consequence will be referred
to below. The calculated MP2 geometries of the C6H5CO and
its transition state are presented in Figure 6 together with those
The second-order rate constants obtained with eq 1 for T <
of the CH CO system for comparison and further discussion
on the effect of resonance stabilization on the thermochemistry
of the benzoyl radical later.
3
4
1
00 K and with eq 7 for T > 400 K are summarized in Table
. A weighted least-squares analysis of these data gives rise to

1236 J. Phys. Chem. A, Vol. 104, No. 6, 2000
Nam et al.
(T,P) given in the table with an adjusted barrier for the
decomposition reaction (E° ). Least-squares analysis of the
-
1
∞
computed k-1 for each mixture using the E° thus derived for
-1
the temperature range 500-1000 K gives the high-pressure
∞
activation energy E , which is also presented in the table.
-
1
∞
1
As revealed by the results presented in the table, E- is
typically greater than the decomposition barrier by about 2 kcal/
mol because of the effect of temperature on the A-factor (or
the partition functions of the TS). Combination of the average
value of E° , 27.3 ( 0.6 kcal/mol, evaluated from the 21 sets
-
1
of Solly and Benson’s data, with the forward reaction barrier,
E1° ) 2.9 ( 0.4 kcal/mol, gives the heat of the decomposition
reaction at 0 K, ∆H° ) 24.4 ( 1.0 kcal/mol.
1
A more reliable alternative approach to evaluate the heat of
reaction is the third-law method, using the equilibrium constant
and the Gibbs energy functions, -(G° - H°)/T, computed with
T
0
the predicted molecular parameters for the reactants (CO and
C6H5) and the product (C6H5CO). The former can be calculated
by the ratio of our forward rate constant presented in Table 1
and that for the reverse reaction calculated at the same
temperature and pressure with the RRKM theory, using E° )
Figure 5. The pseudo-first-order decay plots for the reaction of C
with CO obtained at different total pressures at 347 K. 3, 12 Torr; 9,
0 Torr; 0, 80 Torr; O, 120 Torr. Inset, pressure effect on the second-
order rate constant, k , predicted by the RRKM theory at the temperature
6 5
H
-1
2
7.3 kcal/mol derived above.
4
In the last entry of Table 1, we list the values of ∆H°1
calculated by the third-law method; the average of the values
1
indicated. 500 and 667 K represent the highest temperature used in
5
this and Solly and Benson’s study.
derived by convoluting the all errors of k1 gives ∆H° ) -24.6
1
(
0.1 kcal/mol. Incorporation of the potential error from k-1,
Rate Constant Calculation. Because of the large size of the
present system (8 heavy atoms with open shell), the heat of the
reaction and the barrier for the addition process cannot be
reliably predicted. To circumvent the shortcoming, we calculate
k1 by adjusting the reaction barrier using the RRKM theory with
the computed geometries and vibrational frequencies of the
reactants and the transition state.
In Figure 7 we compare the calculated rate constants using
the molecular parameters computed with the two different
methods. As illustrated in the figure, the RRKM//MP2/6-31G-
as reflected by the deviation in E° ((0.6 as shown in Table
-
1
2
), we arrive at ∆H° ) -24.6 ( 0.7 kcal/mol, which is in
1
close agreement with the value obtained by the first-law method,
H° ) E° - E° ) -24.4 ( 1.0 kcal/mol as presented above.
∆
1
1
-1
The heat of reaction evaluated from the two independent sets
of kinetic data, -24.6 kcal/mol agrees closely with the theoreti-
cally predicted values, -24.6 and -27.6 kcal/mol by B3LYP/
6
-311G(d,p) and MP2/6-31G(d,p), respectively.
In Figure 8, we correlate the forward and reverse rate
∞
constants; the forward rate constant was converted to k by
-
1
(
d,p) calculation with the adjusted E° ) 2.9 kcal/mol appears
1
∞
∞
1
detailed balance (k-1 ) k K ), whereas the reverse rate
-1
to give a better fit to experimental data than the RRKM//B3LYP/
-311G(d,p) result does with the adjusted E° ) 4.0 kcal/mol
constant, listed in Table 2, was extrapolated to the high-pressure
limit with Solly and Benson’s k-1(T,P) as mentioned previously.
The two sets of data can be well correlated with the Arrhenius
expression calculated with the RRKM theory
6
1
(in terms of both temperature dependence and the absolute value
of k1). As indicated above, the latter method predicts a looser
TS with a larger A-factor which requires concomitantly a higher
reaction barrier to match the theory with experiment. The 4.0-
kcal/mol barrier height is greater than the experimental activation
energy by 1 kcal/mol; this is not acceptable and inconsistent
with TST.
∞
-1
14
-1
k ) 5.27 × 10 exp(-14 600/T) s
(10)
using the averaged 0 K barrier E° ) 27.3 kcal/mol and the
-
1
Correlation with the ReVerse Rate Constant. As mentioned
molecular parameters computed by the MP2/6-31G(d,p) method.
5
∞
in the Introduction, Solly and Benson first reported the kinetics
Equation 10 is quite close to Solly and Benson’s value, k
-1
14
-1
for the C6H5CO decomposition reaction using the I2-catalyzed
reaction of C6H5CHO at 614-667 K with pressures varying
from 245 to 461 Torr. Their experimental conditions and the
first-order coefficients, k-1 (T,P), are summarized in Table 2.
Because the mixtures used contain C6H5CHO, I2, HI, and the
key reaction products, CO and C6H5CIO, we listed the averaged
molecular weights ( Mh ) and Lennard-Jones parameters (L-J)
) 4 × 10 exp(-14 800/T) s , obtained by extrapolation with
RRK (taking s ) 20) or RRKM calculations based on assumed
‡
strong collision efficiencies, C H CO structure and vibrational
6
5
5
frequencies.
o
Heat of Formation of C H CO and D (C H -CO). The heat
6
5
0
6
5
of reaction 1, ∆H° ) -24.6 ( 0.7 kcal/mol derived above for
1
0 K, allows us to calculate the heat of formation of the benzoyl
(ꢀ and σ) which are required in our RRKM calculations. Both
radical, ∆ H° ) 32.5 kcal/mol, and ∆ H° ) 29.0 kcal/mol
f
0
f
298
Mh and L-J parameters were averaged with the mole fractions
of individual components. The L-J parameters of C6H5CHO,
I2, HI, and Ar were calculated from their critical constants (Pc,
Tc, and Vc),18 whereas those of C6H5CO and C6H5CIO were
assumed to be the same as C6H5CHO. The L-J parameters were
used for calculation of the effective collision frequencies using
Troe’s weak collision approximation.19
with an approximate total error of (1.5 kcal/mol, using the heats
2
0
21
of formation of C H and CO cited in Table 3. Notably, the
6
5
2
2-24
result agrees well with several reported values,
including
24
the earlier one derived by Solly and Benson given in Table 3.
The lower value of the heat of formation derived by Solly and
Benson, 26.1 at 298 K or 29.6 kcal/mol at 0 K, was obtained
from the kinetics of the I + C H CHO ) HI + C H CO
reaction, log kf (cm molecule s ) ) (13.08 ( 0.16) - (8536
( 180/T), assuming the activation energy for its reverse process
6
5
6
5
∞
3
-1 -1
Table 2 summarizes the high-pressure rate constants (k )
-
1
obtained by RRKM extrapolation of Solly and Benson’s k-1

Kinetics of C6H5 Reaction with CO
J. Phys. Chem. A, Vol. 104, No. 6, 2000 1237
Figure 6. Optimized geometries (given in the left-hand side of the figure) and predicted atomic charge populations and spin densities (numbers
in parentheses), both given in the right-hand side of the figure, for the molecular and transition states of C
MP2/6-31(d,p) method.
6 5 3
H CO and CH CO computed by the

1238 J. Phys. Chem. A, Vol. 104, No. 6, 2000
Nam et al.
TABLE 2: Extrapolation of Solly and Benson’s Data (ref 5)
∞
∞
-1
k
(10
-1(T, P)
E°
k
E
-
1
-1
-4
(kcal mol-1)
T (K)
P (Torr)
Mh (g‚mol-1)
jꢀ (K)
σj (Å)
-4
s
-1
)
(kcal mol-1)
(10 s-1
)
6
14.0
14.0
15.8
15.8
15.8
15.8
30.4
30.9
67.6
67.5
84.3
09.6
09.6
09.7
09.3
09.3
32.5
32.5
32.6
32.5
32.4
66.10
461.30
60.40
55.01
146.62
57.80
167.93
52.82
37.27
47.12
37.70
24.00
29.10
37.10
31.70
59.40
45.00
51.90
38.10
36.40
75.20
169.4
128.1
126.9
118.4
123.2
123.6
114.4
101.2
104.4
112.6
214.2
205.9
240.3
197.9
204.7
195.0
195.4
234.5
194.5
203.8
209.2
386
306
336
316
310
368
298
323
301
301
428
422
438
416
422
419
409
432
410
416
418
5.58
5.50
5.53
5.55
5.53
5.67
5.46
5.56
5.41
5.46
5.70
5.69
5.56
5.68
5.70
5.74
5.61
5.53
5.64
5.64
5.62
1.68
3.00
1.68
2.82
2.86
1.85
3.75
1.82
3.32
4.35
0.70
0.39
0.55
0.74
0.71
0.87
1.12
1.03
1.25
1.32
1.37
26.8
26.6
26.8
26.0
26.4
26.7
26.7
27.3
27.8
27.5
26.3
28.2
27.8
27.5
27.4
27.4
28.0
28.2
27.8
27.7
28.0
27.3 ( 0.6
3.43
3.98
3.58
6.19
4.79
3.89
6.51
4.36
12.01
14.26
1.42
1.08
1.45
1.76
1.77
1.76
2.89
2.56
3.42
3.68
2.97
29.0
28.8
29.0
28.3
28.7
28.9
28.9
29.5
29.8
29.6
28.6
30.2
29.8
29.6
29.6
29.6
30.1
30.2
29.9
29.8
30.0
29.4 ( 0.5
6
6
6
6
6
6
6
6
6
5
6
6
6
6
6
6
6
6
6
6
average
a
a
b
TABLE 3: Summary of Vibrational Frequencies, Moments of Inertia, Relative Energies, and Heats of Formation of
j
Reactants, Products, Transition States for the Reaction of C
6
H
5
6 5
+ CO f C H CO at the MP2/6-31G(d, p) Level of Theory
relative energies (kcal/mol)
∆
f
H° (kcal/mol)
species
I
i
(amu)
u
j
(cm-1)
B3LYP/6-311G(d,p) MP2/6-31G(d,p)
0 K
298.15 K
C
6
H
5
277.462 441, 463, 598, 619, 669, 773, 925, 931,
84.3 ( 0.6c
81.2 ( 0.6c
3
5
11.395 974, 1025, 1038, 1053, 1057, 1076, 1130,
88.857 1179, 1202, 1296, 1494, 1506, 1735, 1786,
079, 3088, 3108, 3113, 3113
3
CO
.000 1999
2.445
2.445
-27.2 ( 0.1d -26.4 ( 0.1d
3
3
C
C
6
H
H
5
+ CO
0.0
-24.58
0.0
-27.55
a
30.3e
27.8e
6
5
CO
318.458 20i , 192, 236, 436, 473, 504, 616, 650,
119.708 701, 808, 855, 955, 1002, 1016, 1092, 1097,
438.166 1137, 1145, 1150, 1184, 1210, 1229, 1323,
f
f
1
1
33.9 ( 1.3
30.6 ( 0.7
26.1 ( 2.0
27.7 ( 2.6
g
h
(29.6 ( 2.0)
(31.2 ( 2.6)
1
3
562, 1596, 1855, 1952, 1976, 3081, 3093,
095, 3109, 3114
333.391 388i, 67, 112, 131, 288, 368, 496, 553, 574,
1384.273 620, 684, 773, 839, 869, 904, 924, 930,
717.664 968, 997, 1096, 1097, 1228, 1250, 1384,
i
29.0 ( 1.5i
32.5 ( 1.5
C
6
H
5
CO
0.73
3.33
(TS)
1
1
2
405, 1458, 1481, 2007, 2963, 2971, 2981,
988, 2991
a
Taken as free rotation case. Includes zero-point energy correction. c Ref 20. d Ref 21. e G2(MP2,SVP) calculation.23 Isodesmic reaction
b
f
2
3
g
h
i
j
calculation.
Ref 24. Ref 22. This work. All frequencies have been scaled by factor 0.9434.
to be 1 ( 1 kcal/mol. More recently, Ng and co-workers23
obtained a higher value of the heat of formation, 33.3 ( 2.2
kcal/mol, from the photofragmentation dynamics study of
acetophenone using a time-of-flight mass spectrometric tech-
nique. This result agrees most closely with our value (32.5 (
1
.5 kcal/mol) and also reasonably with that of Simoes and
Griller, 31.2 ( 2.6 kcal/mol, obtained by photoacoustic calo-
rimetry.22
The heat of reaction for the reverse reaction, (-1), ∆H° )
-
1
o
2
4.6 kcal/mol, represents the bond dissociation energy D0
(
C6H5-CO). This energy is higher than that of the analogous
process,
These resonance structures can lead to a higher stability for
C6H5CO than its alkyl analogue. Solly and Benson,24 on the
other hand, favored solely a contribution from structure II shown
above, with no contributions from structures III-V and thereby
preserving the large resonance energy of the benzene ring.
The above-mentioned argument, however, is not fully sup-
ported by the result of our quantum calculation presented in
Figure 6, in which we compare the bond lengths, atomic charge
CH CO f CH + CO
3
3
o
by 15 kcal/mol (based on the value, D298 (CH3-CO) ) 11
25
kcal/mol ). The greater strength of the >C-CO bond in the
benzoyl radical could be intuitively attributed to the existence
24
of the following resonance structures :

Kinetics of C6H5 Reaction with CO
J. Phys. Chem. A, Vol. 104, No. 6, 2000 1239
spectrometry at temperatures between 295 and 500 K mostly
at 40 Torr Ar pressure. The reaction was weakly pressure-
independent at T > 400 K. Our result could be reasonably
correlated with that of the reverse process reported by Solly
5
and Benson using the RRKM theory with the molecular and
transition-state parameters computed by MP2/6-31G(d,p) cal-
culations. Combination of the two sets of kinetic data yields
the heat of the reaction, C6H5 + CO ) C6H5CO (1), ∆H° )
1
-24.6 kcal/mol by the third-law method. This result, combined
with the known heats of formation of the reactants, gives rise
to the heat of formation of the benzoyl radical, 32.5 ( 1.5 kcal/
mol at 0 K.
The heat of formation of the benzoyl radical thus derived
agrees reasonably with the existing data in the literature within
a rather large range of experimental errors, (3 kcal/mol. Further
work apparently is needed to improve the accuracy.
Figure 7. Comparison of the experimental and theoretically predicted
6 5
results for the C H + CO reaction at the total pressure 40 Torr. O,
experimental result (this work); solid and dashed lines, the RRKM
results calculated with TS parameters obtained by using MP2/6-31G-
(
d,p) and B3LYP/6-311G(d,p) with E° ) 2.9 and 4.0 kcal/mol,
1
Acknowledgment. The authors are grateful for the support
of this work from the Basic Energy Sciences, Department of
Energy, under contract no. DE-FG02-97-ER14784.
respectively. The decrease of the predicted rate constants at higher
temperatures resulted from the falloff effect.
References and Notes
(1) Sawyer, R. F. Twenty-forth Symposium (International) Combustion;
The Combustion Institute: Pittsburgh, PA, 1992; p 1423.
(
(
2) Glassman, I. Combustion, 2nd ed., Academic Press: NY, 1986.
3) Fahr, A.; Mallard, S. G.; Stein, S. E. Twenty-first Symposium
(
1
International) on Combustion; The Combustion Institute: Pittsburgh, PA,
986; p 825.
4) Fahr, A.; Stein, S. E. Twenty-second Symposium (International)
on Combustion; The Combustion Institute: Pittsburgh, PA, 1988; p 1023.
(
(
(
(
(
(
5) Solly, R. K.; Benson, S. W. J. Am. Chem. Soc. 1971, 93, 2127.
6) Yu, T.; Lin, M. C. J. Am. Chem. Soc. 1993, 115, 4371.
7) Lin, M. C.; Yu, T. Int. J. Chem. Kinet. 1993, 25, 875.
8) Yu, T.; Lin, M. C. J. Phys. Chem. 1995, 99, 8599.
9) Yu, T.; Lin, M. C. J. Am. Chem. Soc. 1994, 116, 9571.
Figure 8. Correlation of Benson’s data (O) for C
6
H
5
CO f C
6
H
5
+
(10) Yu, T.; Lin, M. C. Combust. Flame 1995, 100, 169.
(11) Park, J.; Chakraborty, D.; Lin, M. C. J. Phys. Chem. 1999, 103,
003.
(12) Park, J.; Lin, M. C. CaVity-Ring-Down Spectrometry: A New
Technique for Trace Absorption Measurements; ACS Publication Series
20; Washington, DC, 1999, p 196.
13) Diau, E. W. G.; Lin M. C. J. Phys. Chem. 1995, 99, 6589.
14) Becke, A. D. J. Chem. Phys. 1992, 96, 2155; 97, 9173; 1993, 98,
648.
(15) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. 1988, B37, 785.
(16) Mebel, A. M.; Morokuma, K.; Lin, M. C. J. Chem. Phys. 1995,
03, 7414.
CO extrapolated to the high-pressure limit and our rate constant for
∞
4
C
6
H
5
6 5
-1
+ CO f C H CO converted to k (0) with the equilibrium
2
4
3
constant K-1 ) 5.41 × 10 exp(-12 500/T) cm /mol. Solid line, the
predicted first-order rate constant by RRKM calculations (see text).
7
5
1
(
(
populations, and atomic spin densities of C6H5CO with those
of CH3CO. First of all, the dissociating C-C bond in the
benzoyl radical is 1.484 Å, which is shorter than that in CH3-
CO, 1.515 Å. The corresponding C-C bond lengths in the
transition states, however, are reversed, slightly longer in C6H5-
(
17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B. W.; Gill, P. M.;
‡
‡
CO (2.215 Å) than that in CH3CO (2.105 Å), suggesting that
the former is an earlier TS than the latter in the R + CO addition
reactions (reflecting the greater reactivity of C6H5 than CH3).
The charge populations in the carbonyl group of C6H5CO are
only slightly higher than those in CH3CO. Similarly, the spin
density at the C atom of the CO group in C6H5CO, 0.682, is
only slightly lower than that in CH3CO, 0.757, and the
corresponding values at the O atoms, 0.266 and 0.133, respec-
tively, are considerably smaller than what would be expected
for the O atom in structure II, with a value between 1 and 2.
These results indicate that all the resonance structures given
above do contribute partly to the overall greater stability and
thus the higher dissociation energy of the C6H5CO radical.
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
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.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, Revision D.3 ed.;
Department of Chemistry, Carnegie Mellon University, Pittsburgh, PA, 1995.
(
(
(
18) Stiel, L. I.; Thodos, G. J. Chem. Eng. Data 1962, 7, 234.
19) Troe, J. J. Chem. Phys. 1977, 66, 4745.
20) Davico, G. E.; Bierbaum, V. M.; DePuy, C. H.; Ellison, G. B.;
Squires, R. R. J. Am. Chem. Soc. 1995, 117, 2590.
(21) Chase, M. W., Jr.; Davles, C. A.; Downey, J. R., Jr.; Frurip, D. J.;
McDonald, R. A.; Syverud, A. N. JANAF Thermochemical Tables. J. Phys.
Chem. Ref. Data 1985, 14 (Suppl. 1).
(
(
22) Simoes, J. A. M.; Griller, D. Chem. Phys. Lett. 1989, 158, 175.
23) Zhao, H.-Q.; Cheung, Y.-S.; Liao, C.-L.; Ng, C. Y.; Li, W.-K. J.
Concluding Remarks
Chem. Phys. 1997, 107, 7230.
(
(
24) Solly, R. K.; Benson, S. W. J. Am. Chem. Soc. 1971, 93, 1592.
25) CRC Handbook of Chemistry of Physics, 78th ed.; Lide, D. R., Jr.,
In this study, we measured for the first time the rate constant
for the association of C6H5 with CO by cavity ring-down
Ed.; CRC Press: Boca Raton, 1997-98.