Kinetics of the OH + HCNO reaction

Article abstract of DOI:10.1021/jp058305t

The kinetics of the OH + HCNO reaction was studied. The total rate constant was measured by LIF detection of OH using two different OH precursors, both of which gave identical results. We obtain k = (2.69 ± 0.41) × 10 ^-12 exp[(750.2 ± 49.8)/7] cm³ molecule^-1 s^-1 over the temperature range 298-386 K, with a value of k = (3.39 ± 0.3) × 10^-11 cm³ molecule^-1 s^-1 at 296 K. CO, H₂CO, NO, and HNO products were detected using infrared laser absorption spectroscopy. On the basis of these measurements, we conclude that CO + H₂NO and HNO + HCO are the major product channels, with a minor contribution from H₂CO + NO.

Full text of DOI:10.1021/jp058305t

4458
J. Phys. Chem. A 2006, 110, 4458-4464
Kinetics of the OH + HCNO Reaction
Wenhui Feng, Justin P. Meyer, and John F. Hershberger*
Department of Chemistry and Molecular Biology, North Dakota State UniVersity, Fargo, North Dakota 58105
ReceiVed: December 5, 2005; In Final Form: January 24, 2006
The kinetics of the OH + HCNO reaction was studied. The total rate constant was measured by LIF detection
of OH using two different OH precursors, both of which gave identical results. We obtain k ) (2.69 ( 0.41)
× 10-12 exp[(750.2 ( 49.8)/T] cm3 molecule-1 s-1 over the temperature range 298-386 K, with a value of
k ) (3.39 ( 0.3) × 10-11 cm3 molecule-1 s-1 at 296 K. CO, H2CO, NO, and HNO products were detected
using infrared laser absorption spectroscopy. On the basis of these measurements, we conclude that CO +
H2NO and HNO + HCO are the major product channels, with a minor contribution from H2CO + NO.
1. Introduction
This paper also showed ab initio calculations of the OH +
HCNO potential energy surface, predicting pathways to product
Fulminic acid, HCNO, has recently been identified as an
important intermediate in NO-reburning processes for the
reduction of NOx pollutants from fossil-fuel combustion emis-
sions.1 HCNO originates primarily from the oxidation of
acetylene
channels 3a, 3c, 3e, and 3g.1
In this study, we report the first experimental measurements
of the kinetics of the OH + HCNO reaction. In fact, to the best
of our knowledge, this is the first report of any direct
experimental kinetic measurements on fulminic acid, although
a substantial body of literature exists on the kinetics of the more
stable isomer HNCO (isocyanic acid).10-19
O + C2H2 f CO + CH2
f HCCO + H
(1a)
(1b)
(2a)
(2b)
2. Experimental Section
HCCO + NO f HCN + CO2
f HCNO + CO
Total rate constant measurements were measured by laser-
induced fluorescence (LIF) detection of OH. An amount of 193
or 248 nm excimer laser light (typically ∼5 mJ/pulse) was used
to create OH radicals by one of the following two methods
Channel 2b has recently been recognized as the major product
channel in reaction 2, with the most recent measurements and
calculations converging on a branching ratio of æ2a ) 0.22-
0.28.2-6 As a result, the subsequent chemistry of HCNO is of
great interest in the overall NO-reburning mechanism. In
particular, the OH + HCNO reaction has been identified as a
crucial step1
N2O + hν (193 nm) f O(1D) + N2
O(1D) + H2O f OH + OH
(4)
(5)
or
OH + HCNO
f H2O + (NCO) ∆H0298 ) -364.0 kJ/mol (3a)
H2O2 + hν (248 nm) f OH + OH
(6)
f NH2 + CO2 ∆H0298 ) -413.4 kJ/mol
f HNO + HCO ∆H0298 ) -67.1 kJ/mol
f HCN + HO2 ∆H0298 ) -73.0 kJ/mol
f CH2O + NO ∆H0298 ) -235.8 kJ/mol
f CO + H2NO ∆H0298 ) -247.6 kJ/mol
(3b)
(3c)
(3d)
(3e)
(3f)
The excimer light was made collinear with 306.966 nm probe
light from a frequency-doubled Nd:YAG-pumped dye laser.
Both beams entered a 91.4 cm reaction cell. Fluorescence was
detected at right angles from the laser beams with a PMT
mounted against a side window. Signals were recorded on a
gated integrator (with 300 ns delay, 200 ns gate width), digitized,
and stored on a computer. Timing between excimer and dye
pulses was controlled by a computer-controlled digital delay
generator. The pump-probe delay was varied to obtain the OH
concentration profiles.
f CO + H2 + NO ∆H0298 ) -230.4 kJ/mol (3g)
Reaction products were detected by infrared diode laser
absorption spectroscopy. Several lead salt diode lasers (Laser
Components) operating in the 80-110 K temperature range were
used to provide tunable IR probe laser light. The IR beam was
collimated by a lens and combined with the UV excimer light
by means of a dichroic mirror, and both beams were copropa-
gated through a 1.46 m absorption cell. After the UV light was
removed by a second dichroic mirror, the IR beam was then
The thermochemical information has been obtained from
standard tables7 as well as other references for the heats of
formation of HCNO,5 NCO,8 and H2NO.9 In their modeling
study, Miller et al. used an estimated rate constant of k3 ) 3.32
× 10-11 cm3 molecule-1 s-1, with no temperature dependence.1
* To whom correspondence should be addressed. E-mail:
john.hershberger@ndsu.edu.
1
passed into a /4 m monochromator and focused onto a 1 mm
10.1021/jp058305t CCC: $33.50 © 2006 American Chemical Society
Published on Web 03/15/2006

Kinetics of the OH + HCNO Reaction
J. Phys. Chem. A, Vol. 110, No. 13, 2006 4459
InSb detector (Cincinnati Electronics, ∼1 µs response time).
Transient IR absorption signals were recorded on a LeCroy
9310A digital oscilloscope and transferred to a computer for
analysis. To account for small probe laser thermal deflection
effects, signals were collected with the diode laser slightly
detuned off of the absorption lines, and such transients were
subtracted from the on-resonant transients. This was a small
(5-10%) correction for the major products but a more sub-
stantial correction for minor products (NO and H2CO).
SF6 and CF4 (Matheson) were purified by repeated freeze-
pump-thaw cycles at 77 K. Traces of CO2 were removed from
SF6 by the use of an Ascarite trap. NO (Matheson) was puri-
fied by repeated freeze-pump-thaw cycles at 153 K to remove
NO2 and N2O. H2O2 (Aldrich, 50% in H2O) was purified by
extensive pumping to remove the more volatile water compo-
nent. After purification, the H2O2 solution was estimated to be
∼85% pure.
Figure 1. Change in HCNO concentration, monitored by IR absorbance
(using the (01000000) r (0000000) R10 line at 2203.851 cm-1) as a
function of time in the reaction cell. Temperature: 25 °C (diamonds),
115 °C (triangles), and 150° C (crosses).
HCNO was synthesized by vacuum pyrolysis of 3-phenyl-
4-oximino-isoxazol-5-(4H)-one as described in the literature.20-22
The precursor (2.0 g) was sublimed from a 50 mL bulb in a
93-96 °C oil bath and passed through a horizontal quartz tube
heated to 450 °C in a tube furnace. The products, which included
HCNO, HNCO, H2O, CO2, and phenyl cyanide, were collected
over a 12 h period at 77 K. H2O and phenyl cyanide were
removed by twice passing the products through a 240 K trap.
CO2 and HNCO were removed by vacuum distillation at 192
K. A substantial effort was necessary to optimize conditions
for this synthesis. In particular, the 93-96 °C oil temperature
was found ideal to vaporize but not decompose the 3-phenyl-
4-oximino-isoxazol-5-(4H)-one during the pyrolysis. The purity
of the HCNO samples was characterized by FTIR spectroscopy
(especially the comparison of the HCNO absorption at 2196
cm-1 with the HNCO absorption at 2267 cm-1) and was
typically 90% or better. Because HCNO has poor long-term
stability, samples were kept frozen at 77 K except when filling
the reaction cell. Fresh samples were synthesized for each day’s
experiments. In general, HCNO could be allowed to stand at
room temperature for ∼5 min in our Pyrex absorption cell with
minimal decomposition, but significant decomposition over ∼5
min was commonly observed if the sample was in contact with
stainless steel vacuum lines. As a result, the vacuum system
was modified to minimize (but not entirely eliminate) exposure
of HCNO to metal.
molecule measurements, the IR laser beam path was purged
with N2 to remove atmospheric CO2.
Typical experimental conditions were P(HCNO) ) 0.20 Torr,
P(H2O2) ) 0.50 Torr, P(SF6) or P(CF4) ) 1.00 Torr, P(NO) )
0.0-0.50 Torr, and excimer laser pulse energies of ∼4 mJ.
3. Results
3.1. Stability of HCNO Samples. A critical aspect of these
experiments is the requirement that the HCNO samples be
sufficiently stable over the time scale of our experiments. The
samples last about a day in a Pyrex bulb at 77 K, but once
introduced into the reaction cell, some decomposition or
polymerization was inevitable, either due to contact with metal
in the vacuum lines or slow reaction with other reagents in the
cell. To examine this issue, IR diode laser absorption was used
to monitor static HCNO samples as a function of time in the
reaction cell. These data are shown in Figure 1. As is apparent,
only slight decomposition occurs in room temperature over a
3-5 min period, which is the approximate length of time
necessary to collect experimental data. At higher temperatures,
decomposition is significantly faster. On the basis of these
decomposition curves, we choose 386 K as the maximum
temperature to conduct total rate constant experiments. Above
386 K, decomposition occurs sufficiently rapidly enough to
significantly compromise the data.
The following molecules were probed using IR diode laser
absorption spectroscopy
R(14) at 2196.664 cm-1
P(23) at 2202.744 cm-1
3.2. Total Rate Constants. Figure 2 shows a plot of the OH
radical LIF signal as a function of excimer-probe delay time,
using N2O/H2O (reactions 4 and 5) as the OH radical precursor.
In the absence of HCNO reagent, only a very slow decay over
the time range 50-300 µs is observed. This decay is attributed
to the removal of OH radicals by pathways other than the title
reaction, including self-reaction, reaction with other radicals,
and diffusion out of the probed region of the reaction cell. Upon
the addition of HCNO reagent, a large increase in the OH decay
rate is observed. Decay curves such as those shown in Figure
2 were fit to a single exponential over the time range 50-300
µs. Figure 3 shows the resulting decay rates as a function of
HCNO pressure. A linear dependence is observed, as is expected
if the system obeys pseudo-first-order kinetics (for nonzero
CO (V ) 1 r V ) 0)
N2O (0001) r (0000)
CO2 (0001) r (0000)
P(14) at 2337.659 cm-1
HCNO (0100000) r (0000000) R(10) at 2203.851 cm-1
HNO (100) r (000)
P(6) at 2667.785 cm-1
R(10) at 2758.806 cm-1
H2CO (100000) r (000000)
NO (V ) 1 r V ) 0)
R(10.5) at 1912.072 cm-1
The HITRAN molecular database was used to locate and
identify the spectral lines of CO, N2O, H2CO, and CO2 product
molecules.23 Other published spectral data were used to locate
and identify HCNO24 and HNO25 lines. The spectral lines used
are near the peak of the rotational Boltzmann distribution,
minimizing sensitivity to small heating effects. For CO2 product
P
HCNO) and no secondary reactions play a significant role in
the OH signals. As per standard kinetic treatment, the slope of
this plot is the desired rate constant k3. The rate constant k3
obtained in this way was found to be identical within experi-
mental uncertainties for the two OH production methods

4460 J. Phys. Chem. A, Vol. 110, No. 13, 2006
Feng et al.
to obtain reasonably precise Arrhenius parameters. We obtain
the following rate constants
k3(T) ) (2.69 ( 0.41) × 10-12 exp[(750.2 (
49.8)/T] cm3 molecule-1 s-1 (T ) 298-386 K)
k3 ) (3.39 ( 0.3) × 10-11 cm3 molecule-1 s-1 at 296 K
The error bars in the Arrhenius parameters represent one
standard deviation. The error bars in the 296 K value are larger
than the standard deviation of 296 K measurements and are
based primarily on an estimate of the uncertainty in HCNO
concentrations.
The negative activation energy observed in this reaction is
typical of radical-radical kinetics and suggests that the reaction
proceeds by complex formation, followed by possible rear-
rangement and decomposition to form products. Direct hydrogen
abstraction to form H2O + CNO (or NCO) would be expected
to have a significant positive activation energy and is apparently
not a significant contribution in this reaction.
Figure 2. Laser-induced fluorescence signal of OH as a function of
photolysis-probe delay time: circles, 0.0 Torr HCNO; diamonds, 0.0105
Torr HCNO. Other conditions: P(N2O) ) 0.050 Torr, P(H2O) ) 0.350
Torr, P(SF6) ) 0.50 Torr, and excimer at 193 nm.
3.3. Product Yields. IR diode laser absorption was used to
attempt detection of CO, CO2, N2O, NO, CH2O, and HNO
products. Some transient signals are shown in Figure 5. All
product yield data were obtained at 296 K.
CO2 was not successfully detected as a reaction product,
primarily because the HCNO samples contained a trace impurity
of CO2 (even a 1% impurity, difficult to remove, affects our
measurements). In fact, attempts to detect CO2 resulted in small
negative transients, presumably due to removal or vibrational
excitation of CO2 by the chemistry taking place. We did not
investigate this issue at length but simply note that channel 3b
is clearly not a major product channel.
The product molecules CO, NO, CH2O, and HNO were
successfully detected upon 248-nm photolysis of a H2O2/HCNO/
buffer gas mixture. SF6 buffer gas was used for these experi-
ments, with the exception of CO detection, for which CF4 buffer
gas was used. The choice of buffer gas was motivated by the
desire to quickly relax a nascent excited vibrational state
distribution to the Boltzmann distribution. Once this relaxation
is complete, the absolute concentration of the detected product
could be obtained from the population of the probed rotation-
vibration state. Previous measurements in our laboratory6,26-28
as well as vibrational energy transfer measurements29-32 have
demonstrated that SF6 is an effective buffer gas for the relaxation
of vibrationally excited CO2 and N2O as well as many other
molecules but that CF4 is a better buffer gas for relaxation of
vibrationally excited CO.
Figure 3. Pseudo-first-order decay rates of the OH radical as a function
of HCNO pressure. Conditions: P(H2O2) ) 0.50 Torr, P(SF6) ) 0.50
Torr, P(HCNO) ) variable, and excimer at 248 nm.
N2O was detected upon 248-nm photolysis of a H2O2/HCNO/
NO/buffer gas mixture. We attribute N2O formation to the
reaction of NCO with NO
NCO + NO f N2O + CO
f N2 + CO2
(7a)
(7b)
where the branching ratio æ7a ) 0.44 at 296 K.26 As shown
below, however, most or all of the NCO was not produced from
reaction 3 but from secondary chemistry.
HCN is not detectable in our diode laser spectrometer be-
cause of the lack of available laser diodes in the 3300 cm-1
region. As a result, we cannot probe the importance of channel
3d by transient IR spectroscopy. We have, however, obtained
static (Fourier transform) FTIR spectra of the reaction mixture
with and without precursor after extensive photolysis (1000
Figure 4. Arrhenius plot for the OH + HCNO reaction.
(reactions 4 + 5 at 193 nm photolysis wavelength vs reaction
6 at 248 nm).
Figure 4 shows an Arrhenius plot for the OH + HCNO
reaction. As described above, measurements were limited to a
rather narrow temperature range because HCNO decomposition
in the reaction cell becomes unacceptably rapid at high
temperatures. Nevertheless, the temperature range is sufficient

Kinetics of the OH + HCNO Reaction
J. Phys. Chem. A, Vol. 110, No. 13, 2006 4461
Figure 6. Transient signals of (a) H2CO and (b) NO. Reaction
conditions: P(H2O2) ) 0.50 Torr (upper traces), P(H2O2) ) 0.0 Torr
(lower traces), P(HCNO) ) 0.20 Torr, and P(SF6) ) 1.00 Torr.
Transient signal amplitudes (peak-peak) were converted into
absolute concentration using HITRAN linestrengths for CO,
CO2, N2O, CH2O, and NO, as described in previous publica-
tions.26 For HNO, no literature linestrengths were available. As
a result, absolute calibration of HNO signals was done as
follows: HNO and CO transient signals from 248 nm photolysis
of an ICN/H2CO/NO/buffer gas mixture were observed. These
signals are attributed to the following reaction sequence
ICN + hν (248 nm) f I + CN
CN + H2CO f HCN + HCO
(8)
(9)
HCO + NO f HNO + CO
(10)
Figure 5. Transient signals of (a) HNO, (b) CO, and (c) N2O. Reaction
conditions: P(H2O2) ) 0.50 Torr (upper traces), P(H2O2) ) 0.0 Torr
(lower traces), P(HCNO) ) 0.20 Torr, P(SF6) ) 1.00 Torr (HNO and
N2O transients only), P(CF4) ) 1.00 Torr (CO transients only), and
P(NO) ) 0.50 Torr (N2O transients only).
Reactions 9 and 10 are fast, with k9 ) 1.7 × 10-11 cm3
molecule-1 s-1 at 300 K33,34 and k10 ) 1.2 × 10-11 cm3
molecule-1 s-1 at 296 K.35 Provided that an excess of H2CO
and NO are present, we expect quantitative conversion of CN
radicals into HNO + CO products. The HNO concentration may
therefore be assumed to be equal to the CO concentration, which
was obtained from HITRAN linestrengths. In addition, the HNO
(and CO) concentrations should be approximately equal to the
initial number of CN radicals present, which was estimated from
excimer laser pulses). The stable products observed in the
transient experiments are observable in the FTIR, but no HCN
attributable to channel 3d (i.e., beyond that produced by HCNO
photolysis) was detected.

4462 J. Phys. Chem. A, Vol. 110, No. 13, 2006
Feng et al.
TABLE 1: Product Yields of Reaction OH + HCNO
NH + HCNO
f CH2O + N2 ∆H0298 ) -663.7 kJ/mol
excimer
pulse
(13a)
f NH2 + (NCO) ∆H0298 ) -225.6 kJ/mol (13b)
f HCN + HNO ∆H0298 ) -313.1 kJ/mol (13c)
without with
H2O2
energy relative
(mJ)
a
a
product
H2O2 differencea,b
yieldc
NO
H2CO
HNO
0.748
0.0
1.46
1.23
0.376
6.24
16.83
21.11
4.24
0.481
0.376
4.78
10.63
11.68
-0.891
4.56
4.11
3.94
3.80
3.64
5.52
0.033
0.029
0.378
0.872
1.00
O + HCNO
CO (without O2) 6.20
f OH + (NCO) ∆H0298 ) -250.0 kJ/mol (14a)
f CO + HNO ∆H0298 ) -431.4 kJ/mol (14b)
f HCO + NO ∆H0298 ) -286.6 kJ/mol (14c)
CO (with O2)
9.43
5.13
N2Od
a In units of 1012 molecule cm-3 b Obtained by subtracting yields
.
in column 2 from yields in column 3. c Relative yields attributed to
the title reaction, obtained from difference yields (column 4) after
correction for excimer pulse energy (column 5) and normalized to [CO]
) 1.00 d N2O yields obtained in the presence of NO.
The product list in the above reactions is not exhaustive, and
this chemistry at the present time remains almost completely
unexplored, making any detailed modeling of our reaction
system impractical. As an approximate procedure, we have
therefore chosen to simply subtract the product yields obtained
in the absence of H2O2 precursor from the yields obtained with
precursor present to estimate the product yields produced in
the title reaction. This approach is not completely satisfactory,
as H2O2 may somewhat affect the secondary chemistry of some
of the radicals formed in reaction 11. Nevertheless, this
procedure at least allows us to make semiquantitative conclu-
sions regarding the major product channels of the title reaction.
Table 1 shows that the N2O yield upon photolysis of an
HCNO/NO/SF6 mixture is actually somewhat greater than the
yield upon photolysis of an H2O2/HCNO/NO/SF6 mixture. We
attribute the N2O yield obtained from the HCNO/NO/SF6
mixture to the formation of NCO (or possibly CNO) by direct
photolysis, reaction 11a, followed by reaction 7. It is also
possible that some of this N2O originates from photolysis
channel 11c followed by NH + NO secondary chemistry. The
decreased yield obtained when the H2O2 precursor is included
may be due to a reaction of NCO with H2O2, which would
compete with reaction 7, although no literature data exists on
this reaction. In any case, since the N2O yield did not increase
upon including H2O2 in the reaction, we conclude that channel
3a in the title reaction is minor or insignificant. If we assume
that at least a 10% increase in the N2O yield (comparing with
and without H2O2) is the minimum detectable, then we estimate
the photolysis laser energy, the 248 nm absorption coefficient
of the ICN precursor (R ) 0.009 cm-1 Torr-1),26 and an
assumed photolysis quantum yield of unity. These two calibra-
tions gave the same result to within ∼15% uncertainty. The
comparison of the HNO concentration to the HNO transient
signal therefore provided the calibration, which was applied to
the HNO transient signals observed in the title reaction.
Table 1 shows a typical dataset of product molecule yields.
Shown are concentrations obtained both with and without the
H2O2 precursor and the difference between these two measure-
ments. In addition, a small correction was made to account for
variations in excimer laser pulse energies over the course of
the experiment. The right-hand column of Table 1 shows the
resulting relative product yields.
These product yield measurements suffer from one major
background source, which is the direct photolysis of the HCNO
reactants. No detailed study of HCNO photophysics has been
reported, but a UV-vis spectrum of our HCNO samples shows
strong absorption at 193 nm and weaker but still significant
absorption at 248 nm. (It is for this reason that we chose to use
the H2O2 precursor, reaction 6, rather than reactions 4 and 5
for the product yield measurements.) We believe that HCNO
photolysis is probable and is responsible for the observations
of product yields even in the absence of our OH precursor.
Several photolysis products are energetically possible at a 248
nm photolysis wavelength
an upper limit for the branching ratio in channel 3a to be æ3a
<
0.05.
For the other detected products, significant amounts were
detected in the absence of H2O2, but the yields were observed
to increase substantially when H2O2 was included. For the CO,
H2CO, and HNO products, we estimated the contribution of
the title reaction by subtracting the product yield obtained upon
photolysis of HCNO/buffer gas from that obtained upon
photolysis of H2O2/HCNO/buffer gas mixtures. Table 1 shows
the results for a typical experimental run.
Examination of Table 1 shows that the yields of NO and
H2CO are approximately the same. This strongly suggests that
both of these products originate predominately from channel
3e and that channel 3g, which would form an excess of NO, is
comparatively insignificant. The yields of NO and H2CO are,
however, much less than the yields of HNO and CO. This
suggests that channel 3e is in fact a rather minor channel and
that either 3c or 3f dominate the reaction.
HCNO + hν (248 nm) f H + (NCO)
f (HCN) + O
(11a)
(11b)
(11c)
f NH + CO
No information on the relative quantum efficiencies of these
products is available, but it is apparent that the products observed
upon photolysis of a HCNO/SF6 mixture may be formed either
directly by photolysis or by further chemistry of radicals formed
in reaction 11 with other species (including HCNO itself). For
example, we can speculate that perhaps (some) of the following
reactions occur
H + HCNO
f CH2 + NO ∆H0298 ) -87.4 kJ/mol (12a)
Channel 3c could be a source of CO products, because
secondary reactions of HCO could easily form CO. For example,
if trace amounts of oxygen are present
f CO + NH2 ∆H0298 ) -309.4 kJ/mol (12b)
f CH2O + N ∆H0298 ) -32.5 kJ/mol (12c)
HCO + O2 f CO + HO2
(15)

Kinetics of the OH + HCNO Reaction
J. Phys. Chem. A, Vol. 110, No. 13, 2006 4463
TABLE 2: Product Branching Ratios of the OH + HCNO
kinetic, the reaction of OH with transient species produced by
HCNO photolysis is clearly not a severe problem. If k3 were
several orders of magnitude slower, such secondary chemistry
would be a much more serious problem. A more problematic
issue is the reaction of OH with stable products, which could
potentially build up over the ∼50 laser shots required to obtain
one of the data sets of Figure 2. A simple test for this effect is
to simply measure the OH decay twice in rapid succession on
the same gas fill. If the OH reaction with stable products were
a significant problem, we would expect the second measurement
to yield a higher value of the pseudo-first-order OH decay rate.
In fact, we observe only a slight decrease in this decay rate on
the second measurement, consistent with a slight decomposition
of HCNO over time.
The mechanism of this reaction likely proceeds from OH
attack at the carbon of HCNO to form a HC(OH)NO complex.
Hydrogen atom migration via a four-center transition state can
form a HC(O)N(H)O complex, which may then dissociate by
C-N bond fission to form HCO + HNO, channel 3c.
Alternatively, this complex may undergo a second hydrogen
atom migration via a three-center transition state to form an
OCNH2O complex, which may dissociate to CO + H2NO
products, channel 3f. The first of these possibilities, channel
3c, was predicted in the potential energy surface of Miller et
al.1 That study did not consider the possibility of channel 3f,
however. That study also predicted a low-energy pathway to
H2 + CO + NO products, channel 3g. Our observation of very
low NO yields suggests that this is not a major channel.
Pathways to channel 3a involve two hydrogen migrations to
the terminal oxygen of HCNO. The second of these migrations
involves a moderately higher barrier, but the authors still
predicted a moderate yield of NCO-forming channels.1 Our
experiments indicate that formation of NCO is at most a very
minor pathway in this reaction. Pathways to other product
channels, such as 3d, were predicted to have high energy
barriers.
Reactiona
product channel
branching fraction
CO + H2NO (3f)
HNO + HCO (3c)
NO + H2CO (3e)
0.61 ( 0.06
0.35 ( 0.06
0.04 ( 0.02
a Assumes no other product channels are active.
could occur. Other HCO secondary chemistry may be possible
as well, along with simple HCO f H + CO dissociation. In
some experiments, we included O2 as a reagent in the reaction
mixture, to convert any HCO formed into CO via reaction 15.
As shown in Table 1, this resulted in a moderate increase in
the observed CO yield. This result certainly suggests that at
least some of the observed CO originates from HCO, however,
the CO yield is much in excess of the HNO yield, so channel
3c cannot account for all of the observed CO. We believe that
the CO yield obtained in the presence of excess O2 is in fact
measuring the contribution of CO from channel 3f as well as
the HCO from 3c. We estimate the contribution of 3f by
subtracting the HNO yield from the CO yield (with O2 included)
and conclude that 3f is in fact the major channel of the title
reaction. Table 2 shows the estimated branching ratios obtained,
where we have assumed that 3c, 3e, and 3f are the only active
product channels (i.e., the totals are normalized to unity). The
total yield of 3c + 3e + 3f is consistent with an order of
magnitude estimate of [OH]0 ∼ 1013 molecules cm-3, but we
cannot exclude the possibility that other product channels may
exist as well. Although different data sets give very consistent
results to within approximately 2%, the error bars in Table 2
are somewhat greater to account for possible systematic errors,
considering the potential secondary chemistry described above.
4. Discussion
Our results represent the first experimental study of the title
reaction. The only literature comparison is an estimated total
rate constant for modeling purposes and a computational study
of the OH + HCNO potential energy surface at the HL1 level
of theory.1 Our total rate constant at 296 K is nearly exactly
the value of 3.32 × 10-11 cm3 molecule-1 s-1 used in the
modeling study. That study, however, made no estimate of
temperature dependence. An extrapolation of our measurements
to the temperature range T ) 1200-1500 K relevant in NO-
reburning suggests a rate constant nearly an order of magnitude
below our 296 K value, although clearly such an extrapolation
is not warranted by the narrow temperature range of our data.
Several experimental artifacts can cause systematic errors in
a pseudo-first-order kinetics experiment. The first, decomposi-
tion of HCNO samples during the experiments, was minimized
by completing a single OH LIF decay measurement in about 4
min (typically, this allows 3 min for filling the cell and allowing
the reagents to mix and 1 min for the LIF data collection). In
this amount of time, less than 10% decomposition occurs, as
shown in Figure 1. A related issue is the possible reaction of
OH radicals with decomposition products, photoproducts, or
reaction products from the title and/or secondary reactions. By
using an estimated absorption cross section of HCNO at 248
nm of 1.41 × 10-19 cm2, and assuming a photolysis quantum
yield of unity, we estimate ∼6 × 1012 cm-3 of HCNO photolysis
products from a 5 mJ excimer pulse at PHCNO ) 0.06 Torr (the
highest pressure used in Figure 3). This is comparable to our
estimated values of [OH]0, but represents only approximately
0.3% conversion of the initial HCNO. Because our measured
rate constants are within an order of magnitude of the gas
5. Conclusion
The kinetics and product branching of the OH + HCNO
reaction were studied using laser-induced fluorescence and IR
diode laser absorption spectroscopy. The reaction is fast, with
k ) (3.39 ( 0.3) × 10-11 cm3 molecule-1 s-1 at 296 K, and
has a moderate, negative temperature dependence. The major
product channels are CO + H2NO and HCO + HNO.
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.
References and Notes
(1) Miller, J. A.; Klippenstein, S. J.; Glarborg, P. Combust. Flame 2003,
135, 357.
(2) Boullart, W.; Ngugen, M. T.; Peeters, J. J. Phys. Chem. 1994, 98,
8036.
(3) Eickhoff, U.; Temps, F. Phys. Chem. Chem. Phys. 1999, 1, 243.
(4) Vereecken, L.; Sumathy, R.; Carl, S. A.; Peeters, J. Chem. Phys.
Lett. 2001, 344, 400.
(5) Tokmakov, I. V.; Moskaleva, L. V.; Paschenko, D. V.; Lin, M. C.
J. Phys. Chem. A 2003, 107, 1066.
(6) Meyer, J. P.; Hershberger, J. F. J. Phys. Chem. B 2005, 109, 8363.
(7) Chase, M. W., Jr. NIST-JANAF Thermochemical Tables, 4th ed;
Journal of Physical and Chemical Reference Data; American Chemical
Society: Washington , DC, 1998.
(8) East, A. L. L.; Allen, W. D. J. Chem. Phys. 1993, 99, 4638.
(9) Sumathi, R.; Sengupta, D.; Nguyen, M. T. J. Phys. Chem. A 1998,
102, 3175.
(10) Tsang, W. J. Phys. Chem. Ref. Data 1992, 21, 753.
(11) Miller, J. A.; Melius, C. F. Int. J. Chem. Kinet. 1992, 24, 421.

4464 J. Phys. Chem. A, Vol. 110, No. 13, 2006
Feng et al.
(12) Mertens, J. D.; Hanson, R. K. Symp. (Int). Combust. [Proc.] 1996,
26, 551.
(22) Wilmes, R.; Winnewisser, M. J. Labelled Compd. Radiopharm.
1993, 33, 157.
(13) Nguyen, M. T.; Sengupta, D.; Vereecken, L.; Peeters, J. Vanquick-
enborne, L. G. J. Phys. Chem. 1996, 100, 1615.
(14) Mertens, J. D.; Kohse-Hoinghaus, K.; Hanson, R. K. Bowman, C.
T. Int. J. Chem. Kinet. 1991, 23, 655.
(15) Wooldridge, M. S.; Hanson, R. K.; Bowman, C. T. Int. J. Chem.
Kinet. 1996, 28, 361.
(16) Mertens, J. D.; Change, A. Y.; Hanson, R. K.; Bowman, C. T. Int.
J. Chem. Kinet. 1992, 24, 2479.
(17) Tully, F. P.; Perry, R. A.; Thorne, L. R.; Allendorf, M. D. Symp.
(Int.) Combust. [Proc.] 1989, 22, 1101.
(18) Ongstad, A. P.; Liu, X.; Coombe, R. D. J. Phys. Chem. 1988, 92,
5578.
(19) Liu, X.; Machara, N. P.; Coombe, R. D. J. Phys. Chem. 1991, 95,
4983.
(20) Pasinszki, T.; Kishimoto, N.; Ohno, K. J. Phys. Chem. 1999, 103,
6746.
(21) Wentrup, C.; Gerecht, B.; Briehl, H. Angew. Chem., Int. Ed. Engl.
1979, 18, 467.
(23) Rothman, L. S.; et al. J. Quant. Spectrosc. Radiat. Transfer 1992,
48, 469.
(24) Ferretti, E. L.; Rao, K. N. J. Mol. Spectrosc. 1974, 51, 97.
(25) Johns, J. W. C.; McKellar, A. R. W.; Weinberger, E. Can. J. Phys.
1983, 61, 1106.
(26) Cooper, W. F.; Park, J.; Hershberger, J. F. J. Phys. Chem. 1993,
97, 3283.
(27) Cooper, W. F.; Hershberger, J. F. J. Phys. Chem. 1992, 96, 771.
(28) Meyer, J. P.; Hershberger, J. F. J. Phys. Chem. A 2005, 109, 4772.
(29) Fakhr, A.; Bates, R. D., Jr. Chem. Phys. Lett. 1980, 71, 381.
(30) Stephenson, J. C.; Moore, C. B. J. Chem. Phys. 1970, 52, 2333.
(31) Richman, D. C.; Millikan, R. C. J. Chem. Phys. 1975, 63, 2242.
(32) Green, W. H.; Hancock, J. K. J. Chem. Phys. 1973, 59, 4326.
(33) Yu, T.; Yang, D. L.; Lin, M. C. Int. J. Chem. Kinet. 1993, 25,
1053.
(34) Chang, Y.-W.; Wang, N. S. Chem. Phys. 1995, 200, 431.
(35) Tsang, W.; Herron, J. T. J. Phys. Chem. Ref. Data 1991, 20, 609.