Propargyl from the Reaction of Singlet Methylene with Acetylene

Article abstract of DOI:10.1021/jp952147w

The technique of infrared kinetic spectroscopy has been used to study the production of propargyl radical from the reaction of singlet methylene with acetylene.The rate constant for this product channel was determined to be (3.5 +/- 0.7)E-10 cm³ molecule^-1 s^-1 at 295 K, measured relative to the known rate for ¹CH2 with H2 or CH4.Methylene was produced in the singlet state by excimer laser photolysis of ketene at 308 nm in the presence of acetylene and either H2 or CH4.Reaction of ¹CH2 with acetylene produces propargyl, and reaction of ¹CH2 with either H2 or CH4 produces CH3.The intensity of a propargyl infrared absorption line was compared with that of a methyl infrared absorption line, and the rate of formation of propargyl was determined from the ratio of these two intensities and the known rate of reaction of singlet methylene with H2 (or CH4) to produce CH3.The relative peak infrared absorption cross sections of methyl and propargyl were calibrated under the conditions of the experiment by photolyzing crotyl bromide at 193 nm to produce methyl and propargyl in equal concentrations.

Full text of DOI:10.1021/jp952147w

J. Phys. Chem. 1996, 100, 2125-2128
2125
Propargyl from the Reaction of Singlet Methylene with Acetylene
J. D. Adamson, C. L. Morter, J. D. DeSain, G. P. Glass,* and R. F. Curl*
Department of Chemistry and Rice Quantum Institute, Rice UniVersity, Houston, Texas 77251
X
ReceiVed: July 25, 1995; In Final Form: October 30, 1995
The technique of infrared kinetic spectroscopy has been used to study the production of propargyl radical
from the reaction of singlet methylene with acetylene. The rate constant for this product channel was
-
10
3
-1 -1
determined to be (3.5 ( 0.7) × 10 cm molecule
s
at 295 K, measured relative to the known rate for
1
2 2 4
CH with H or CH . Methylene was produced in the singlet state by excimer laser photolysis of ketene at
08 nm in the presence of acetylene and either H
1
3
2
4
or CH . Reaction of CH
2
with acetylene produces propargyl,
1
and reaction of CH
2
with either H
2
or CH
4
produces CH
3
. The intensity of a propargyl infrared absorption
line was compared with that of a methyl infrared absorption line, and the rate of formation of propargyl was
determined from the ratio of these two intensities and the known rate of reaction of singlet methylene with
H (or CH ) to produce CH . The relative peak infrared absorption cross sections of methyl and propargyl
2 4 3
were calibrated under the conditions of the experiment by photolyzing crotyl bromide at 193 nm to produce
methyl and propargyl in equal concentrations.
Introduction
bonded complexes with acetylene should dominate the equi-
1
librium. Thus, they argue that propargyl recombination is the
more plausible mechanism for aromatic ring formation. They
propose that propargyl is formed in flames by the reaction
sequence:
Soot formation in combustion appears to be tied closely to
the formation of polycyclic aromatic hydrocarbons (PAH). If
aromatic rings are already present, plausible mechanisms for
the addition of carbon species to a ring resulting in new rings
and the growth of PAH have been proposed. However, for a
flame burning an aliphatic fuel, the mechanism by which the
initial aromatic ring is formed is still unclear. The current
situation concerning this issue is well summarized in a recent
paper by Miller and Melius.1
C H + O f HCCO + H
(R1)
(R2)
(R3)
2
2
1
HCCO + H f CH + CO
2
1
Flame studies undertaken during the mid-1980s led several
CH + C H f C H (propargyl) + H
2
2
2
3
3
2
-5
groups
to propose that benzene is formed by either the
2
6
reaction of C4H5 with acetylene or the reaction of C4H3 with
In experimental studies, Westmoreland et al. have shown that
3
4,5
acetylene or both. These mechanisms have been summarized
sufficient C species are found in acetylene flames to account
3
6
and extended by Westmoreland et al. However, Kern and co-
for aromatic ring formation (the C ’s are presumably formed
3
7
workers have suggested that benzene produced in the pyrolysis
by the mechanism above) but that this is not the case for 1,3-
of allene and 1,3-butadiene is formed by the recombination of
propargyl radicals, and Alkemade and Homann have studied
propargyl recombination directly and observed formation of
several C6H6 isomers, one of which is benzene. More recent
studies by Kern and co-workers on benzene production in the
high temperature decomposition of propargyl bromide conclude
that it is the result of a sequence of reactions initiated by the
recombination of C3 species.
In their analysis of likely mechanisms for soot formation in
flames, Miller and Melius1 make the point that at high
temperatures rapid isomerization can occur as a result of
hydrogen atom exchange reactions, and thus chemical equilib-
rium between the isomers of C4H5 and C4H3 is rapidly
established. The isomers that can form strong bonds upon
condensation with acetylene are the less stable normal radicals,
butadiene flames. They suggest that both these C and C
3
4
8
mechanisms occur.
Previous studies on the reaction of singlet methylene with
acetylene have determined overall rate constants for the reac-
9
10
-10
3
-1 -1
11
tion of (3.7 ( 0.3) × 10
cm molecule
s
and (2.93
-10
3
-1 -1
(
0.19) x 10
cm molecule s , but the branching ratio
into the propargyl product channel has not previously been
determined. In this work, the propargyl radical was directly
observed via infrared kinetic spectroscopy as a product of the
1
reaction of CH2 with C2H2 and the rate constant for reaction 3
determined.
Experimental Section
The infrared kinetic spectrometer has been previously de-
12
scribed. Briefly, the unfocused beam of an excimer laser is
overlapped with an IR color center laser probe beam in a
multipass (White) flow cell. For rate determinations, singlet
methylene was produced by photolysis of ketene at 308 nm
using a XeCl excimer laser operated at 10 Hz with 108 mJ/
pulse. At this wavelength, the photolysis products have been
HCtCsCHdCH
(n-C H )
4 3
H CdCsCtCH T H CdCdCdCH
2
2
(
i-C H )
4 3
1
13, 14
demonstrated to be predominantly CH2 and CO.
In the
H CdCHsCHdCH
2
experiment, diketene (Aldrich) was entrained in a flow of helium
and ketene produced by pyrolyzing the diketene stream in a
quartz tube heated to 600 °C which was then directly introduced
into the flow cell as the impurities were determined not to affect
our experiments. Methane (99.9%), hydrogen (99.999%), and
helium buffer gas (99.995%) were used without further purifica-
tion. Partial pressures of various reagents are given in Table
1. Acetone was removed from acetylene by passage through a
(
n-C H )
4 5
H CdCsCHdCH T H CdCdCHsCH
2
2
2
2
(
i-C H )
4 5
but the resonantly stabilized iso radicals which form only weakly
X
Abstract published in AdVance ACS Abstracts, January 1, 1996.
0022-3654/96/20100-2125$12.00/0 © 1996 American Chemical Society

2
126 J. Phys. Chem., Vol. 100, No. 6, 1996
Adamson et al.
TABLE 1: Partial Pressures in the Flow Cell during
Kinetics Measurements
species
partial pressure (Torr)
C
2
H
2
1.25-1.34
1.21-4.32
1-2
CH
CH
He
4
or H
CO
2
2
10.0
total
14.3-16.4
copper coil immersed in a dry ice/acetone bath. The methyl
signal was monitored at 3154.7468 cm-1 (N′ ) 2, K′ ) 1 r
15
N ) 2, K ) 0), and the propargyl R(8) transition of the ν1 CH
1
6
-1
stretch was monitored at 3327.8484 cm .
For calibration purposes, propargyl and methyl were produced
by photolysis of crotyl bromide (Aldrich) at 193 nm using an
ArF excimer laser operated at 10 Hz with 100 mJ/pulse under
the conditions used in the kinetics experiments. In order to
avoid interferences, as will be described later in this paper, d2-
acetylene was synthesized from calcium carbide (Aldrich) and
deuterium oxide (99.9% Aldrich), acetylene was synthesized
1
Figure 1. Time profile of CH
and without (dashed line) acetylene in the presence of CH
2 2
of CH :C H of 2:1. The spike at t ) 0 is an artifact associated with
3
at 3154.7468 cm- with (solid line)
4
in a ratio
4
the excimer laser discharge.
from calcium carbide and water, and d4-diketene was synthe-
sized from d6-acetone (99.9% Cambridge Isotopes).1
7,18
The
1
purity and identity of the d4-diketene were checked via H NMR
1
3
and C NMR.
Kinetics
Singlet methylene reacts too quickly (<1 µs) to be observed
on the normal time scale of our apparatus. Therefore, direct
measurement of the rate of formation of propargyl by following
the appearance of propargyl would have required special
measures to improve the detector time response. Moreover,
propargyl is probably produced in excited vibrational states and
must be quenched before its concentration can be reliably
obtained from measurements on the ground state. Under these
reaction conditions, complete vibrational quenching in less than
1
Figure 2. Time profile of C
3
H
3
produced from the reaction of CH
2
-1
+
2
2 2 4 4
C H at 3327.8484 cm in the presence of CH in a ratio of CH :
C H of 2:1. This figure corresponds with Figure 1. The spike at t )
0 is an artifact associated with the excimer laser flash. The ripples on
the decay curve are believed to be caused by reflected shock waves
2
(from the cell walls) from the photolysis.
1
µs is problematical. Therefore, the rate of (R3) was measured
by competition using the following reactions that produce
methyl, which we can observe:
are given in Table 1. Note that in Figure 1 the CH3 time profile
with and without C2H2 demonstrates the competition between
reaction R3 and reaction R4 or R5.
Over about the first 200 µsec, the decay curves of Figures 1
and 2 are determined entirely by reactions removing CH3 (Figure
1
CH + H f CH + H
(R4)
(R5)
2
2
3
1
CH + CH f 2CH
3
2
4
1
) or C3H3 (Figure 2) as diffusion of even the lighter CH3 takes
1
place on a multi-millisecond time scale. The decay curves
which Figures 1 and 2 exemplify cannot be fitted with an
exponential form. Efforts to combine first order and second
order kinetics to fit the decays did not result in believable kinetic
parameters. Therefore, the initial C3H3 and CH3 column
densities were obtained through back-extrapolation of ap-
proximately the first 30-60 µs after the flash of time profiles
such as those in Figures 1 and 2. The principal hazard of such
treatment is that population is still funneling into the state
observed during the extrapolation period. A hint of a situation
like this is seen in the initial approximately 20 µs after the flash
in the trace without C2H2 in Figure 1 where the signal continues
to rise after the flash. Note that that behavior is completely
absent in the trace with C2H2 in Figure 1 probably because C2H2
is an effective quencher of vibrationally excited CH3. There is
also no indication of a continued rise after the flash in the much
noisier signal of Figure 2. Because there is no lingering rise
signal in either of the cases actually measured, we believe that
this simple extrapolation measurement treats the data adequately.
The other reagents are present in large excess to CH2 so that
pseudo-first-order kinetics apply. Integration of the appropriate
pseudo first order rate equations yields
k [H ][C H ]
3 0
4
2
3
k3 )
(1)
[
C H ][CH ]
2 2 3 0
or
2
k [CH ][C H ]
5 4 3 3 0
k3 )
(2)
[
C H ][CH ]
2 2 3 0
where [C3H3]0 and [CH3]0 are the concentrations of these species
after all CH2 has been consumed, but before any decay through
subsequent reactions of these radicals has occurred. The rate
1
10
constant for reaction R4 has been measured: k4 ) (1.2 ( 0.3)
-
10
3
-1
1
×
10
cm s . The overall rate of removal of CH2 by
1
9
-11
3
-1
methane is (7 ( 0.4) × 10
cm s , and the quenching
1
3
20
-11
3
rate of CH2 to CH2 by methane is (1.2 ( 0.4) × 10
s , making k5 about 6 × 10 cm s . The rate of reaction
R3 was determined for three different ratios of [C2H2]:[H2] and
C2H2]:[CH4] of 1:1, 1:2, and 1:3. Similar results were obtained
for each set of conditions. Typical pressures in the flow cell
cm
-1
-11
3
-1
Calibration
[
In order to calculate the CH3 and C3H3 column densities,
absorption cross sections for the specific transitions used must

Reaction of Singlet Methylene with Acetylene
J. Phys. Chem., Vol. 100, No. 6, 1996 2127
be known. In previous work,12 cross sections had been
ening caused by the deuterated species can be assumed very
nearly identical with that produced by the normal species. In
these calibrations, only mixtures duplicating the H2 kinetic
measurements were used, since the CH4 containing mixtures
will still result in a CH3 being produced (R5). Calibration was
completed through comparison of the CH3 transition at 3154.7468
-
1
determined for a C3H3 transition at 3314.704 cm and a CH3
-1
transition at 3247.4863 cm . These could not be used in these
kinetics measurements, however, due to interference from
acetylene absorptions in these regions. After comparing
potential C3H3 and CH3 transitions to known C2H221 transi-
tions, two candidates that did not overlap with acetylene peaks
were found. For C3H3, the R(8) transition at 3327.8484 cm-
was chosen, and for CH3 the N′ ) 2, K′ ) 1 r N ) 2, K ) 0
16
15
-1
-1
cm to the C3H3 transition at 3327.8484 cm under reaction
conditions. The resulting ratio between the two cross sections
was σCH :σC H ) 2.77:1.
1
3
3 3
-
1
transition at 3154.7468 cm was chosen. These transitions
were then monitored throughout the kinetics experiments with
the excimer laser running at 308 nm.
Blank Checking Experiments
Several experiments were carried out to check that the signals
observed were not contaminated by artifact sources of the
radicals observed produced by some sort of side reaction. In
the kinetics work at 308 nm we looked for, and did not observe,
CH3 with the following variations of conditions: diketene
pyrolysis mixture (ketene) alone, methane alone, C2H2 alone,
ketene + C2H2. At 308 nm, we looked for, and did not observe,
C3H3 with ketene alone.
Since the cross sections for a CH3 peak and a C3H3 peak had
previously been determined, our first approach to determining
the cross sections for the two new lines was to simply compare
the new lines to the old calibrated lines and determine relative
cross sections by comparing absorbances. This simplistic
approach, however, was not implemented because pressure
broadening was found to affect both lines (3327.8484 and
-1
3
154.7468 cm ) under the conditions used in the kinetics
In the calibration work at 193 nm, the d4-diketene was
checked by running it through the cell and looking for CH3
and C3H3. Neither was observed.
measurements, and it cannot be assumed that the change in the
peak absorption cross section is the same for the two lines
actually used in the rate measurements.
The integrated intensity of a line is not affected by pressure
broadening. However, actual integration of a peak requires a
very high signal-to-noise and a very flat base line as the line
wings are quite extensive. We considered our signals did not
meet either criterion. For a single line, the measurement of
the frequency width (full width at half-maximum) and peak
intensity assuming a Voigt profile can be used to obtain reliable
integrated intensity and we have sufficient signal-to-noise for
such measurements. However, all our lines have underlying
and nearby substructure that prevents application of a Voigt
profile model. Therefore, we found it necessary to calibrate
peak absorption coefficients under conditions identical with
those of our observations.
Results
Once the CH3 and C3H3 transitions monitored during the
kinetics experiments were calibrated, the rate constant for
reaction 3 was calculated. First, the following ratio was
evaluated for each set of conditions:
[
CH or H ]A
4
2
C H ,0
3
3
R )
(3)
[C H ]A
2
2
CH ,0
3
where Ax,0 indicates infrared absorbance at t ) 0. The Beer-
Lambert law can be expressed
In previous, unrelated, unpublished work in our lab, we
discovered that photolysis of crotyl bromide (1-bromo-2-butene)
yielded propargyl, which indicates that it should also yield
methyl. It was first verified that CH3 and C3H3 are indeed
produced in a concentration ratio of 1:1 by using the old,
A ) σcL
(4)
where A is the base e absorbance, σ is the absorption cross
section, c is the concentration, and L is the path length. Due to
the complex nature of the overlap region between the UV
photolysis beam and the IR probe beam, L is not known
accurately, but when the ratio of absorbances is taken, it cancels
out. Therefore, to determine a value for k3, the ratios R were
averaged, multiplied by the ratio of the cross sections of the
CH3 and C3H3 peaks, r.
-1
previously calibrated lines (3314.704 and 3247.4863 cm ),
carrying out the photolysis under the conditions used in their
original line calibration. Thus, a viable method of calibrating
the ratio of the line peak absorption cross sections under the
conditions of the experiments would seem to be to photolyze
crotyl bromide at 193 nm, with all the constituents of the kinetic
system present to account for pressure broadening. Unfortu-
r ) σCH3/σ
(5)
C H
3
3
1
nately, ketene is still photolyzed at 193 nm producing CH2
-
10
cm molecule s-1, since
3
-1
and then multiplied by 1.2 × 10
and HCCO + H.2
2
-10
cm molecule s-1 and k5 ) 0.5 k4 (see eqs
3
-1
k4 ) 1.2 × 10
-10
1
1 and 2). This results in a value for k3 of (3.5 ( 0.7) × 10
3 -1 -1
CH CO f CH + CO
(R6)
(R7)
2
2
cm molecule s . The values for R under varying concentra-
tion ratios are given in Table 2.
CH CO f HCCO + H
2
Discussion
1
Since there is production of CH2 at 193 nm, there will also be
conversion to extra CH3 upon addition of H2 or CH4 and reaction
with C2H2 to produce extra C3H3. Interference by the methyl
and propargyl produced by reactions R3 and R4 could be
Such a large value for k3 compared to the two reported overall
1
0
-10
3
-1
rate constants for the reaction of 3.7 × 10 cm molecule
-
1
11
-10
3
-1 -1
1
s
and 2.93 × 10
cm molecule
s
between CH2 and
C2H2 indicates that this product channel accounts for most of
avoided by using C2D2 and D2. However, in addition, H reacts
1
0
with ketene to produce methyl.10
the reaction. In the study which produced the higher rate,
the quenching rate of CH2 by acetylene was measured
1
-
11
3
-1 -1
CH CO + H f CH + CO
(R8)
independently to be (8 ( 3) × 10 cm molecule s . Thus,
2
3
1
this study would say that the CH2 reaction rate with quenching
-10
3
-1 -1
In order to avoid all these unwanted sources of CH3 and C3H3,
deuterated diketene and C2D2 were used. The pressure broad-
omitted is 2.9 × 10
cm molecule s , which is less than
-10
3
-1 -1
our measured value of (3.5 ( 0.7) × 10 cm molecule
s

2
128 J. Phys. Chem., Vol. 100, No. 6, 1996
TABLE 2: Ratios Under Varying Concentration Ratios
3 3
2 2 2
:CH :C R ) ([CH or H ]AC H ,0/([C H ]ACH3,0)
Adamson et al.
(3) Frenklach, M.; Clary, D. W.; Gardiner, W. C.; Stein, S. E. 20th
Symp. (Int.) Combust. 1984, 887. Frenklach, M.; Warnatz, J. Combust. Sci.
Technol. 1987, 51, 265.
H
2
4
2
H
2
4
(
4) Colket, M. B. 21st Symp. (Int.) Combust. 1987, 851.
(5) Harris, S. J.; Weiner, A. M.; Blint, R. J. Combust. Flame 1988,
2, 91.
(6) Westmoreland, P. R.; Dean, A. M.; Howard, J. B.; Longwell, J. P.
0
1
0
2
0
3
:1:1
:0:1
:2:1
:0:1
:3:1
:0:1
1.01
0.96
1.06
1.10
1.14
1.10
7
J. Phys. Chem. 1989, 93, 8171.
(7) Wu, C. H.; Kern, R. D. J. Phys. Chem. 1987, 91, 6291. Kern, R.
D.; Singh, H. J.; Wu, C. H. Int. J. Chem. Kinet. 1988, 20, 731.
av 1.06 ( 0.07a
(8) Alkemade, U.; Homann, K. H. Z. Phys. Chem. (Munich) 1989, 161,
) (3.5 ( 0.7) × 10 cm molecule s-1
-
10
3
-1
resulting k
3
1
9.
a
(9) Kern, R. D.; Chen, H.; Kiefer, J. H.; Mudipalli, P. S. Combust.
Flame 1995, 100, 177.
10) Baulch, D. L.; Cobos, C. J.; Cox, R. A.; Esser, C.; Frank, P.; Just,
Th.; Kerr, J. A.; Pilling, M. J.; Troe, J.; Walker, R. W.; Warnatz, J. J. Phys
Chem. Ref. Data 1992, 21, 411.
It is possible to combine the data for H
average in this manner only because 2k ) k within experimental error.
Note there are no systematic differences in R between H and CH
2 4
and CH into a single
5
4
(
2
4
.
for the reaction producing propargyl. When all the measurement
uncertainties are taken into account, there is probably no
disagreement between any of the three studies, although our
results are in better agreement with the older, higher value of
the singlet methylene plus acetylene rate. Since our rate
constant for the formation of propargyl is as high as the rate
(
11) Hayes, F.; Gutsche, G. J.; Lawrance, W. D.; Staker, W. S.; King,
K. D. Combust. Flame 1995, 100, 653.
(12) Morter, C. L.; Farhat, S. K.; Adamson, J. D.; Glass, G. P.; Curl, R.
F. J. Phys. Chem. 1994, 98, 7029.
(13) Hayden, C. C.; Neumark, D. M.; Shobatake, K.; Spark, R. K.; Lee,
Y. T. J. Chem. Phys. 1982, 76, 3607.
(
(
14) Kelley, P. M.; Hase, W. L. Chem. Phys. Lett. 1975, 35, 57.
15) Amano, T.; Bernath, P. F.; Yamada, C.; Endo, Y.; Hirota, E. J.
1
constant for loss of CH2 by collision with acetylene, it appears
that the formation of propargyl is the only pathway resulting in
chemical reaction.
Chem. Phys. 1982, 77, 5284.
16) Morter, C. L.; Domingo, C.; Farhat, S. K.; Cartwright, E.; Glass,
G. P.; Curl, R. F. Chem. Phys. Lett. 1992, 195, 316.
(
(
(
(
17) Williams, J. W.; Krynitsky, J. A. Org. Syn. Collect. 1932, 3, 508.
18) Williams, J. W.; Hurd, C. D. J. Org. Chem. 1940, 5, 122.
19) Langford, A. O.; Petek, H.; Moore, C. B. J. Chem. Phys. 1983, 78,
Acknowledgment. This work was supported by grants from
the Department of Energy and the Robert A. Welch Foundation.
We also thank Christoph Gesenberg and Professor Edward
Billups for their invaluable help in the synthesis of d4-diketene.
6
650.
(20) B o¨ hland, T.; Temps, F.; Wagner, G. Gg. Ber. Bunsen-Ges. Phys.
Chem. 1985, 89, 1013.
References and Notes
(21) Lafferty, W. J., private communication.
(22) Unfried, K. G.; Glass, G. P.; Curl, R. F. Chem. Phys. Lett. 1991,
(
1) Miller, J. A.; Melius, C. F. Combust. Flame 1992, 91, 21.
2) Cole, J. A.; Bittner, J. D.; Howard, J. B.; Longwell, J. P. Combust.
1
77, 33.
(
Flame 1984, 56, 51.
JP952147W