Initial observations of ketene in flow reactor kinetic studies

Article abstract of DOI:10.1524/zpch.2001.215.8.1011

Variable Pressure Flow Reactor (VPFR) experiments were conducted for ethene/methane mixtures utilizing both FTIR analyses of extracted gas samples and in situ line-of-sight far-UV absorption measurements. In situ observations of ketene prompted the re-evaluation of the presence of ketene in previous acetaldehyde oxidation and pyrolysis experiments in the VPFR. In these experiments only extractive sampling (with FTIR analysis) had been used and ketene had not been observed. In reconsidering the previously obtained FTIR spectra, ketene is shown to be not only present, but also present in levels on the order of other important acetaldehyde reaction intermediates, such as CH₂O, H₂, and C₂H₄. Through a comparison of the results of the in situ and extractive ketene diagnostics, it has been shown that significant losses of ketene can occur in sample handling and that the principal ketene loss mechanism appears to be surface dominated. It is likely that this difficulty in ketene sampling, common to many experimental systems, has led to the omission of important ketene pathways in many kinetic models. by Oldenbourg Wissenschaftsverlag, Muenchen.

Full text of DOI:10.1524/zpch.2001.215.8.1011

Zeitschrift für Physikalische Chemie, 215, 8, 1011–1023 (2001)
by Oldenbourg Wissenschaftsverlag, München
Initial Observations of Ketene in Flow Reactor
Kinetic Studies
∗
By James J. Scire, Jr., Stephen D. Klotz and Frederick L. Dryer
Department of Mechanical and Aerospace Engineering, Princeton University, Princeton,
New Jersey 08544
(Received August 11, 2000; accepted September 22, 2000)
Ketene / Acetaldehyde / Flow Reactor / In situ Measurements /
Infrared Spectra
Variable Pressure Flow Reactor (VPFR) experiments were conducted for ethene methane
mixtures utilizing both FTIR analyses of extracted gas samples and in situ line-of-sight
far-UV absorption measurements. In situ observations of ketene prompted the re-evaluation
of the presence of ketene in previous acetaldehyde oxidation and pyrolysis experiments in
the VPFR. In these experiments only extractive sampling (with FTIR analysis) had been
used and ketene had not been observed. In reconsidering the previously obtained FTIR
spectra, ketene is shown to be not only present, but also present in levels on the order
of other important acetaldehyde reaction intermediates, such as CH2O, H2, and C2H4.
Through a comparison of the results of the in situ and extractive ketene diagnostics, it
has been shown that significant losses of ketene can occur in sample handling and that
the principal ketene loss mechanism appears to be surface dominated. It is likely that
this difficulty in ketene sampling, common to many experimental systems, has led to the
omission of important ketene pathways in many kinetic models.
Introduction
Ketene (CH2CO) is thought to be an important intermediate in hydrocarbon
combustion, yet its reactivity makes it difficult to sample and quantify. In fact,
ketene is rarely reported as an intermediate in experimental kinetic studies, and
studies of the high-temperature kinetics of ketene as a parent fuel have been
confined to shock tubes (e.g. [1, 2]). In studies where ketene is reported as an
intermediate, researchers have noted that its high reactivity makes rapid sample
analysis necessary [3], and the difficulty of maintaining pure ketene samples
for spectroscopic use has been discussed by numerous researchers (e.g. [4]).
These factors demonstrate the difficulty in experimentally studying the role of
ketene as an intermediate species in hydrocarbon combustion.
*
Corresponding author. E-mail: fldryer@princeton.edu

1012
J. J. Scire, Jr., St. D. Klotz and F. L. Dryer
In the course of a recent study by our group on the oxidation of
ethene methane mixtures in a variable pressure flow reactor, ketene was de-
tected in situ. This detection prompted closer examination of our extractive
samples for ketene, and in subsequent experiments, ketene was detected both
extractively and in situ. Discrepancies between the in situ and extractive meas-
urements motivated this study, which, in part, examines the loss of ketene in
our sample system for certain ethene methane experiments. Prior to this study,
we would not have expected ketene to survive the extractive sampling and sub-
sequent handling used in our experiments. However, in the present study the
ketene loss mechanism is characterized, and it is now evident that for many
of our flow reactor experiments, including those conducted prior to this study,
ketene loss from the sample is small. More importantly, we are now able to
examine the role of ketene as an intermediate in kinetic systems important to
hydrocarbon combustion. In our ethene methane oxidation studies, this capa-
bility allows us to examine the importance of various channels in the reaction
of vinyl radicals with molecular oxygen. There is considerable disparity in
literature rate constants for these channels (e.g. [5, 6]). Here, this capability al-
lows us to re-examine previous experiments on acetaldehyde kinetics, where
the role of ketene has been especially controversial.
The role of ketene in acetaldehyde kinetics
Acetaldehyde (CH3CHO) is among the major intermediates formed during the
combustion of many hydrocarbons. In particular, it is a major reaction interme-
diate formed during the oxidation of ethanol [7]. Unfortunately, acetaldehyde is
a very reactive species under conditions that produce photochemical smog [8]
and is among the toxic compounds found in automotive exhaust, as defined by
the U. S. Clean Air Act [9], and heavily targeted for reduction. As a result of
these factors, acetaldehyde has been studied extensively through experiments
and kinetic modeling over the last 25 years. Despite the attention acetaldehyde
has received, there still remains considerable uncertainty and disagreement
over the relative importance of its various consumption pathways, particularly
for those involving ketene.
Early work by Rice and Herzfeld [10] resulted in the development of a ki-
netic mechanism for acetaldehyde pyrolysis. This mechanism contained the
following major steps:
CH3CHO ↔ CH3 +CHO
HCO ↔ H+CO
(1)
(2)
(3)
(4)
(5)
(6)
H+CH3CHO ↔ H2 +CH3CO
CH3 +CH3CHO ↔ CH4 +CH3CO
CH3CO ↔ CH3 +CO
CH3 +CH3 ↔ C2H6

Initial Observations of Ketene in Flow Reactor Kinetic Studies
1013
Acetaldehyde was consumed primarily by thermal decomposition to form
methyl (CH3) and formyl (CHO) radicals and by methyl radical abstraction
of the aldehydic hydrogen of acetaldehyde to form methane and acetyl radi-
cals (CH3CO). The possibility also existed (and still exists) for methyl radicals
to abstract alkyl hydrogens from acetaldehyde, a pathway that would result in
the formation of ketene. However, a wealth of acetaldehyde experimental data
and modeling studies consistently suggested this alkyl hydrogen abstraction to
be highly unlikely, since ketene was not previously observed in acetaldehyde
oxidation and pyrolysis experiments.
The Rice and Herzfeld mechanism was used later in an acetaldehyde ex-
perimental study by Liu and Laidler [11]. Acetaldehyde pyrolysis data were
obtained over the 480–540 ◦C temperature range and at pressures from 1 to
560 mm Hg. In order to model the experimental results, the following ketene
formation steps were added to the Rice and Herzfeld mechanism:
CH3 +CH3CHO ↔ CH4 +CH2CHO
CH2CHO ↔ CH2CO+H
(7)
(8)
The primary reason for including these steps, however, was not to model ketene
(no ketene was detected), but rather to provide a possible pathway for hydrogen
production. In fact, subsequent work from this same group [12] reported that
ketene was not formed during low temperature acetaldehyde pyrolysis and it
was concluded that the Liu and Laidler mechanism required modification. The
modification consisted of removing the ketene formation step (reaction 8) and
including
CH2CHO+CH3CHO ↔ CH3CHO+CH3CO
(9)
CH2CHO+CH3CHO ↔ CH3COCH = CHOH+H
(10)
The first flow reactor study of acetaldehyde, by Colket et al. [13], also
failed to observe ketene, even though it was speculated to form under the
high-temperature conditions of the study ( 1000 ◦C). Through a detailed an-
alysis of the mechanism and comparison with their acetaldehyde pyrolysis
data, Colket and co-workers showed that the primary consumption route for vi-
noxy radicals was reaction 8 at the higher temperatures. Benson also reached
this conclusion in his analysis of acetaldehyde pyrolysis [14]. Colket and co-
workers attributed their failure to observe ketene to their sample storage and
gas analysis procedures, after failing to detect ketene in acetone pyrolysis,
where it is known to be a major product. Thus, although ketene was not ob-
served, ketene pathways were retained in their modeling. In a similar fashion,
a stirred reactor study of acetaldehyde kinetics failed to detect ketene, but
the experimental results were modeled with a kinetic mechanism containing
a ketene sub-mechanism [15].
Current modeling of acetaldehyde continues to rely heavily upon the prin-
cipal acetaldehyde consumption pathways depicted earlier in the Rice and

1014
J. J. Scire, Jr., St. D. Klotz and F. L. Dryer
Herzfeld mechanism [11]. The relative unimportance of the alkyl hydrogen ab-
straction pathway continues to be supported by the lack of evidence of ketene
in acetaldehyde oxidation and pyrolysis experiments. In fact, among the kinetic
data reported in The NIST Chemical Kinetics Database [16], the only rate data
available for reaction 7 are from Laidler and Liu [17], and the only rate data
for reaction 8 are from Colket et al. [13]. The fact that these rates have not
been updated more recently most likely stems from the fact that this secondary
pathway has continued to be regarded as unimportant.
The flow reactor experiments of Colket et al. were recently revisited and
extended to higher pressures by Zarubiak [18]. Unlike in the earlier experi-
ments of Colket et al., continuous sampling was performed, and on-line Fourier
transform infrared spectrometry of the flowing sample was used to quantify
products and intermediates. Since this technique allows the entire mid-infrared
spectrum (4000 to 600 cm−1) to be rapidly scanned, and since the spectra
obtained in all previous experiments were retained electronically, additional
analyses of the archival data were possible. In the present work, ketene is quan-
tified in Zarubiak’s experiments to demonstrate the important role of ketene in
acetaldehyde kinetics.
Experimental
In both the earlier experiments of Zarubiak [18] and in the present experiments
on ethene methane oxidation, data were obtained using a Variable Pressure
Flow Reactor (VPFR). The reactor (Fig. 1) is described in considerable de-
tail elsewhere [18, 19]. In the reactor, pre-heated nitrogen carrier gas is mixed
with oxygen and proceeds into a 10 16 cm diameter by 1 m long quartz reac-
tor duct. The carrier gas oxygen mixture flows around an 8 9 cm quartz faced
baffle plate into a 0 64 cm gap serving as the entrance to a silica foam diffuser
with a 5 degree half angle. The remaining reactants (fuel, additives) are diluted
with nitrogen and injected radially between the baffle plate and the diffuser, op-
posite the direction of carrier gas flow (which flows radially inward), allowing
for rapid mixing.
After fuel injection and mixing, the reacting mixture proceeds through the
diffuser and reactor duct past the sampling position. At the sampling position,
a hot water cooled, stainless steel sampling probe continuously extracts and
convectively quenches a small portion of the flow at the centerline of the reac-
tor. The sample gas flows via heated teflon lines to a set of analytical equipment
including a Fourier transform infrared spectrometer (FTIR), an electrochemical
O2 analyzer, and a pair of non-dispersive infrared analyzers for CO and CO2
(for further details, see [19]). The FTIR used for the ethene methane experi-
ments has a resolution of 0 5 cm−1 and is equipped with an 8 6 m multi-pass
cell. The cell has an internal volume of 1 5 L, and is maintained at 373 K and
780 Torr. For the acetaldehyde experiments, an earlier FTIR, with a resolution

Initial Observations of Ketene in Flow Reactor Kinetic Studies
1015
Fig. 1. Schematic of the variable pressure flow reactor used in this study. Also shown is
the experimental configuration for in situ absorption measurements in the far-UV.
of 0 656 cm−1, was used with the same cell. The temperature of the reacting
mixture is measured at the sampling position with a silica-coated type R ther-
mocouple. In addition to the sample probe and thermocouple, the reactor vessel
also houses windows to allow for optical access at the sampling location. The
distance between the point of fuel injection and the sampling position is varied
by moving the injector with attached mixer diffuser assembly relative to the
fixed sampling location by means of a stepper-motor driven slide table. Mean
velocity measurements along the centerline of the reactor have been used to
correlate distance with residence time. The dilute nature of the reacting mix-
tures ensures that the maximum temperature rise due to reaction heat release
is only a small percentage of the initial gas temperature. Five individually con-
trolled electrical resistance heaters maintain the local wall temperature of the
quartz reactor duct at the initial gas temperature, thereby approximating an
adiabatic reaction zone.
In the ethene methane experiments reported here, line-of-sight far-UV ab-
sorption measurements were made via the optical access windows. In the opti-
cal system (Fig. 1), a collimated beam from a deuterium lamp is split and the
two resulting beams are mechanically chopped at different frequencies. One

1016
J. J. Scire, Jr., St. D. Klotz and F. L. Dryer
beam is directed through the reactor, passing once through the reactor duct.
This beam is then recombined with the second beam, which serves as a refer-
ence. The combined beams are dispersed by a monochromator and detected by
a photomultiplier tube. The measurement and reference signals are then recov-
ered by separate lock-in amplifiers monitoring the chopper frequencies. Full
details of the optical configuration appear in [20].
Ketene diagnostics
The strongest infrared band of ketene is the CCO antisymmetric stretch at
2152 cm−1. Measurements of this band are complicated by the strong absorp-
tion by carbon monoxide around 2143 cm−1. The overlap of these spectra
hindered Zarubiak in his attempts to identify ketene in acetaldehyde experi-
ments, and given the historical absence of ketene in acetaldehyde experiments
and modeling, no special attention was given to the ketene formation routes.
Nevertheless, as the rotational spacing for ketene is much smaller than for CO
(0.3 versus 1 9 cm−1), and as the ketene spectrum shows separation of its K
sub-branches, it is possible to distinguish the two if the CO mole fractions are
not too great. Thus, any ketene present can be quantified from the FTIR spectra
(including those previously obtained by Zarubiak).
CO and ketene were thus measured simultaneously by a non-linear least
squares fit of calculated spectra to the experimental FTIR spectra. In the fit-
ting procedure, transmission spectra for given mole fractions of CO and ketene
are initially calculated at high resolution (with 0 030 cm−1 data spacing). The
results are then Fourier transformed and truncated with an apodization func-
tion to simulate the FTIR spectrum. This experimental resolution spectrum is
compared to the actual FTIR spectrum, allowing for a baseline shift. For the
high resolution calculations, the ketene absorption spectrum is obtained by in-
terpolating an experimental spectrum. Absolute ketene intensity was inferred
from ketene destruction experiments (to be discussed with the results). Since
the CO lines are narrower than the FTIR resolution, the CO absorption spec-
trum is simulated, using literature values for the absolute band intensity [21],
rotational constants [22], and collisional half-widths [23] with the rigid rotor
intensity expressions for the line strengths [24]. The fitting procedure yields ex-
cellent results for CO calibration data from 14 to 4400 ppm, with a correlation
coefficient of 0.99924 for a linear regression between the fit results and the ac-
tual values. False ketene response for these files was present for the high CO
files but was generally small, with a maximum false reading of 4 ppm ketene
for the 4400 ppm CO file.
For the present experiments, in situ ketene measurements were made by
monitoring absorption in the electronic band between 195 nm and 218 nm,
using the oscillator strength reported by Rabalais et al. [25]. Our measurements
are slightly complicated at high temperatures, given the lack of published
high-temperature ketene spectra. Without knowledge of hot band positions and

Initial Observations of Ketene in Flow Reactor Kinetic Studies
1017
intensities, the quantification requires that the absorption be integrated over the
band, and accounting for these corrections, we estimate the uncertainty in our
measurements to be 20%.
Results and discussion
As mentioned in the introduction, ketene was initially identified in a high-
pressure ethene methane oxidation experiment, where it was detected in situ.
For this same experiment, however, ketene did not appear in the FTIR spec-
tra. Initially, it was assumed that, as with past studies, this species was too
reactive to survive in the sample system. However, subsequent experiments
clearly showed ketene in the FTIR spectra. Because the experiment where
ketene was not observed in the FTIR had much higher reactant mole fractions
than those where ketene was observed in the FTIR, it seemed likely that the
sample composition was affecting the rate of ketene loss in the sample system.
We therefore examined the effects of sample composition on ketene destruction
in our sampling system.
Ketene destruction in the VPFR sampling system
The pressure and temperature of the sample gas extracted from the VPFR drop
rapidly as the gas passes through the sample probe. The flow through the probe
is frictionally choked, so the flowrate is fixed by the reactor pressure and op-
erating temperature. The majority of the sample gas is released to exhaust,
while a small portion is directed through the sample train. In the experiments
conducted here, at pressures and temperatures in the non-explosive kinetic
regime [26], the radical termination reactions that occur in the quenching pro-
cess have minimal effect on the stable species concentrations in the sample
gas [27]. As such, the destruction of ketene is more likely to have occurred
after the sampling probe, in the low temperature (∼ 373 K) sample transfer
lines and the FTIR cell. The average FTIR cell residence time is 27 seconds,
for a sample flowrate of 2.5 SLPM. In order to observe ketene destruction, the
sample flowrate was varied from 0.9 to 3.25 SLPM, while the initial sample gas
composition was fixed. Initially, the sample gas originated from C2H4 CH4 ox-
idation at 6 atm for initial conditions of 960 K, 0 47% O2, 0 92% C2H4, 0 39%
CH4 at a fixed reactor residence time of 0 70 s. Subsequently, additional experi-
ments were conducted with the initial oxygen mole fraction doubled.
As the cell residence time was increased (smaller flowrate), there was
a clear decrease in the amount of ketene measured extractively, while the con-
centrations of other species (such as C2H4, CH4, O2, and H2O) were essentially
constant. Subtraction of the spectra for cases with different cell residence times
yields the spectra of species involved in the ketene conversion reactions. In
the subtraction spectrum shown in Fig. 2, the ketene spectrum is clearly vis-
ible at 2152 cm−1 (with some remnants of the CO lines). At 1796 cm−1 the

1018
J. J. Scire, Jr., St. D. Klotz and F. L. Dryer
Fig. 2. Subtraction of two spectra for the same reaction residence time at two different
sample flowrates (3.23 SLPM minus 0.98 SLPM). Reaction is at 6 atm with initial condi-
tions 960 K, 0 47% O2, 0 92% C2H4, 0 39% CH4 at a residence time of 0 70 s.
inverted spectrum of acetic acid is visible. Examination of the remainder of the
spectrum (from 4000 cm−1 to 600 cm−1) revealed the remaining features of the
acetic acid spectrum, including the strong bands at 1184 and 643 cm−1 and the
weaker band at 3581 cm−1. With the exception of a small change (∼ 20 ppm)
in CO2, which can be traced to a slight variation in the reactor conditions, no
spectra, other than those attributable to ketene and acetic acid, were found to
appear as differences.
Additional subtraction spectra were obtained for different sample flowrate
differences. Using these spectra, the acetic acid was quantified using existing
calibration spectra and the ketene was measured as described above. For these
initial calculations an integrated extinction coefficient of 500 km mol for the
ketene was used. A plot of the ketene lost versus the acetic acid gained, shown
in Fig. 3, illustrates that the relationship is nearly linear with a negligible off-
set. As we have been unable to find experimental measurements of the band
intensity for ketene, the data shown in this plot were used to adjust the band
intensity to yield a one-to-one correspondence between the acetic acid produc-
tion and ketene destruction. Ketene measurements made with this calibration
provide a lower limit to the actual ketene value, as some ketene may be lost to
other species with much lower infrared intensities that were not detectable on
the FTIR. In addition, the reaction producing the acetic acid (described below)
is most likely heterogeneous, so the acid has some potential to be retained on

Initial Observations of Ketene in Flow Reactor Kinetic Studies
1019
Fig. 3. Conversion of ketene to acetic acid in the sample system.
the surface. It seems, however, that the production of acetic acid is the dom-
inant process, since the in situ ketene measurements matched the sum of the
FTIR ketene and acetic acid measurements within experimental error.
From our results, we conclude that the ketene loss in the sampling system
arises primarily from the reaction:
CH2CO+H2O → CH3COOH
(11)
This reaction has received considerable theoretical study (e.g. [28]). It has
been found to involve hydrogen-bonded water clusters in the gas phase, and to
proceed through a 1,1-enediol intermediate [28]. The gas phase reaction was
experimentally studied by Blake and Davies [29] between 496 and 597 K in
a static silica reactor, where it was found that, without the catalytic action of
acetic acid already present in the vessel, the reaction was heterogeneous. Our
C2H4 CH4 experiments showed that for low water concentrations, the rate of
ketene loss increased with increasing water. Between 3000 and 6000 ppm of
water, however, there was only a small increase in the loss rate. This satura-
tion effect is consistent with a surface dominated loss mechanism. In addition,
in some cases with the same sample gas composition and flowrates, different
rates of ketene destruction were found. This “memory” effect is also consis-
tent with a surface dominated loss mechanism. Then, given the small residence
time in the teflon sample lines, the experimental evidence of this study suggests

1020
J. J. Scire, Jr., St. D. Klotz and F. L. Dryer
that the conversion of ketene to acetic acid is primarily occurring on the nickel
surface of the FTIR cell.
Acetaldehyde pyrolysis and oxidation
From the above results it was clear that ketene may very well have survived
long enough to appear in the spectra from the acetaldehyde work of Zaru-
biak [18]. Many of the cases in that study had water levels below 3000 ppm,
and were conducted with the same sample flowrate and cell conditions as those
used in the C2H4 CH4 experiments. Closer examination of the archival spec-
tra from Zarubiak’s experiments, for both the oxidation and the pyrolysis of
acetaldehyde, in fact, revealed ketene.
Two of Zarubiak’s cases were re-analyzed to quantify ketene. The first case
was an oxidation experiment at 950 K and 5 atm, with an initial acetaldehyde
Fig. 4. Acetaldehyde oxidation with initial conditions of 950 K, 5200 ppm CH3CHO,
2 5, at a pressure of 5 atm. For clarity, the CH2O, H2, and C2H4 data are not shown, but
=
note that CH2O, H2, and 10×C2H4 profiles are similar to that of C2H6.

Initial Observations of Ketene in Flow Reactor Kinetic Studies
1021
mole fraction of 5200 ppm and an equivalence ratio of 2.5. Partial data for this
experiment are shown in Fig. 4. We note immediately that the ketene is of the
same order as the other important intermediates, CH2O, C2H6, and H2. Second,
in acetaldehyde pyrolysis at 1000 K and 5 atm, with 5600 ppm of fuel, the ob-
served level of ketene is even greater, with a maximum ketene mole fraction
of 125 ppm, while in that experiment the ethane only reaches 100 ppm. It is
clear that ketene plays an important role in the kinetics of acetaldehyde. As it
is present in both the oxidation and the pyrolysis, it is unlikely that ketene is
being generated from acetic acid. Rather, it seems more likely that it is being
generated from vinoxy decomposition (reaction 8). The experimental observa-
tions and associated analysis of this study therefore support the importance of
hydrogen abstraction from the methyl group of acetaldehyde, as originally pro-
posed by Liu and Laidler, a reaction channel that has been largely neglected in
acetaldehyde modeling.
Using an existing mechanism for dimethyl ether oxidation [30], which
includes acetaldehyde reactions as a sub-mechanism, we compared model
predictions with Zarubiak’s pyrolysis case just discussed. The modeling pre-
dictions were obtained using several variations on the mechanism, including
various combinations of the estimates from Colket et al. [13], the estimates
from Benson [14], and the ketene chemistry of Hidaka et al. [1]. In each
case, the ketene mole fractions were under-predicted by over a factor of two.
The species profiles of other species, especially C2H4 and C2H6, were also
in considerable disagreement. The modeling results make it clear that con-
siderable work, beyond the scope of this paper, will be required to bring
current acetaldehyde mechanisms into accord with these new experimental
observations.
Conclusions
It has been shown, through examination of in situ and extractive measure-
ments on ethene methane mixture oxidation, that significant losses of ketene
can occur in sample handling. For our system, the loss mechanism appears
to be surface dominated. It is likely that this difficulty in ketene sampling,
common to many experimental systems, has led to the omission of important
pathways in many kinetic models. In particular, review of acetaldehyde data
has shown ketene to be an important intermediate with mole fractions on order
of those of ethane in both the oxidation and pyrolysis of this fuel. This result
indicates that the often-neglected step of hydrogen abstraction from the methyl
group of acetaldehyde is indeed important. The consistent failure to experi-
mentally measure ketene is reflected in the poor ketene predictions of current
acetaldehyde models. It has been demonstrated here that experiments utiliz-
ing on-line, continuous sampling can help to minimize ketene sampling losses.
Additional experimental work on acetaldehyde kinetics, incorporating these

1022
J. J. Scire, Jr., St. D. Klotz and F. L. Dryer
techniques along with in situ measurement techniques for ketene, will help to
clarify this important area in hydrocarbon kinetics. These techniques are also
important to other studies where ketene is an important intermediate, such as
our current work examining vinoxy production from the reaction of vinyl with
molecular oxygen.
Acknowledgements
This work, including the initial acetaldehyde experiments, was supported by
the U.S. Department of Energy, Chemical Sciences Division, Office of Ba-
sic Energy Sciences, under Contract # DE-FG02-86ER-13503. We would also
like to thank Mr. Paul Michniewicz for his assistance in conducting the ex-
periments and Prof. Richard Yetter and Dr. Andrei Kazakov for their thoughtful
discussions.
References
1. Y. Hidaka, K. Kimura and H. Kawano, Combust. Flame 99 (1994) 18.
2. Y. Hidaka, K. Kimura, K. Hattori and T. Okuno, Combust. Flame 106 (1996) 155.
3. P. F. Nelson and M. J. Wornat, Energy & Fuels 4 (1990) 415.
4. W. F. Arendale and W. H. Fletcher, J. Chem. Phys. 26 (1957) 793.
5. J. W. Bozzelli and A. M. Dean, J. Phys. Chem. 97 (1993) 4427.
6. P. R. Westmoreland, Combust. Sci. and Tech. 82 (1992) 151.
7. T. S. Norton and F. L. Dryer, Int. J. Chem. Kinet. 24 (1992) 319.
8. W. P. L. Carter and R. Atkinson, Environ. Sci. Tech. 23 (1989) 864.
9. CAAA – Clean Air Act Amendment (1990). Title 42, U. S. Code, Chap. 85, Section
7545.
10. F. O. Rice and K.F. Herzfeld, J. Am. Chem. Soc. 56 (1934) 284.
11. M. T. H. Liu and K. J. Laidler, Can. J. Chem. 46 (1968) 479.
12. H. P. Schuchmann and K. J. Laidler, Can. J. Chem. 48 (1970) 2315.
13. M. B. Colket, D. W. Naegeli and I. Glassman, Int. J. Chem. Kinet. 7 (1975) 223.
14. S. W. Benson, Thermochemical Kinetics, John Wiley & Sons: New York (1976).
15. P. Dagaut, M. Reuillon, D. Voisin and M. Cathonnet, Combust. Sci. Technol. 107
(1995) 307.
16. W. G. Mallard, F. Westley, J. T. Herron, R. F. Hampson and D. H. Frizzell, The NIST
Chemical Kinetics Database (1998).
17. K. J. Laidler and M. T. H. Liu, Proc. Roy. Soc. London A 297 (1967) 365.
18. D. C. Z. Zarubiak, Masters Thesis, Department of Mechanical and Aerospace En-
gineering, Princeton University, Princeton, NJ (1997).
19. T. J. Held, Ph. D. Thesis, Department of Mechanical and Aerospace Engineering,
Princeton University, Princeton, NJ (1993).
20. J. J. Scire, Jr., Ph. D. Thesis, Department of Mechanical and Aerospace Engineer-
ing, Princeton University, Princeton, NJ, in preparation (2001).
21. P. L. Varghese and R. K. Hanson, J. Quant. Spectrosc. Radiat. Transfer 24 (1980)
479.
22. K. P. Huber and G. Herzberg, Constants of Diatomic Molecules, Van Nostrand Rein-
hold: New York (1979).

Initial Observations of Ketene in Flow Reactor Kinetic Studies
1023
23. J. M. Hartmann, L. Rosenmann, M. Y. Perrin and J. Taine, Appl. Opt. 27 (1988)
3063.
24. G. Herzberg, Spectra of Diatomic Molecules, 2nd ed., D. Van Nostrand, New York
(1950).
25. J. W. Rabalais, J. M. McDonald, V. Scherr and S. P. McGlynn, Chem. Rev. 71 (1971)
73.
26. J. J. Scire, Jr., R. A. Yetter and F. L. Dryer, Int. J. Chem. Kinet. 33 (2001) 75.
27. J. J. Scire, Jr., R. A. Yetter and F. L. Dryer, Joint Meeting of United States Sections:
The Combustion Institute, Washington, D. C., March 15–17 (1999), p. 188.
28. M. T. Nguyen and G. Raspoet, Can. J. Chem. 77 (1999) 817.
29. P. G. Blake and H. H. Davies, J. Chem. Soc. Perkin Trans. II (1972) 321.
30. S. L. Fischer, F. L. Dryer and H. J. Curran, Int. J. Chem. Kinet. 32 (2000) 713.