Reflected shock tube studies of high-temperature rate constants for OH + NO2 -HO2 + NO and OH + HO2 ± H 2O + O2

Article abstract of DOI:10.1021/jp057461x

The motivation for the present study comes from the preceding paper where it is suggested that accepted rate constants for OH + NO₂ + NO → HO₂ are high by ～2. This conclusion was based on a reevaluation of heats of formation for HO₂, OH, NO, and NO₂ using the Active Thermochemical Table (ATcT) approach. The present experiments were performed in C₂H₅I/NO₂ mixtures, using the reflected shock tube technique and OH-radical electronic absorption detection (at 308 nm) and using a multipass optical system. Time-dependent profile decays were fitted with a 23-step mechanism, but only OH + NO₂, OH + HO ₂, both HO₂ and NO₂ dissociations, and the atom molecule reactions, O + NO₂ and O + C₂H₄, contributed to the decay profile. Since all of the reactions except the first two are known with good accuracy, the profiles were fitted by varying only OH + NO₂ and OH + HO₂. The new ATcT approach was used to evaluate equilibrium constants so that back reactions were accurately taken into account. The combined rate constant from the present work and earlier work by Glaenzer and Troe (GT) is k_OH+NO2 = 2.25 × 10 ^-11 exp(-3831 K/T) cm³ molecule^-1 s ^-1, which is a factor of 2 lower than the extrapolated direct value from Howard but agrees well with NO + HO₂ → OH + NO₂ transformed with the updated equilibrium constants. Also, the rate constant for OH + HO₂ suitable for combustion modeling applications over the T range (1200-1700 K) is (5 ± 3) × 10¹¹ cm³ molecule^-1 s^-1. Finally, simulating previous experimental results of GT using our updated mechanism, we suggest a constant rate for k _HO2+_NO2 = (2.2 ± 0.7) ± 10^-11 cm³ molecule^-1 s 1 over the T range 1350-1760 K.

Full text of DOI:10.1021/jp057461x

6602
J. Phys. Chem. A 2006, 110, 6602-6607
Reflected Shock Tube Studies of High-Temperature Rate Constants for OH + NO2 f HO2
†
+ NO and OH + HO2 f H2O + O2
Nanda K. Srinivasan,‡ Meng-Chih Su,‡,§ James W. Sutherland,‡,| Joe V. Michael,*,‡ and
Branko Ruscic‡
Chemistry DiVision, Argonne National Laboratory, Argonne, Illinois 60439-4831
ReceiVed: December 22, 2005; In Final Form: February 9, 2006
The motivation for the present study comes from the preceding paper where it is suggested that accepted rate
constants for OH + NO2 f NO + HO2 are high by ∼2. This conclusion was based on a reevaluation of heats
of formation for HO2, OH, NO, and NO2 using the Active Thermochemical Table (ATcT) approach. The
present experiments were performed in C2H5I/NO2 mixtures, using the reflected shock tube technique and
OH-radical electronic absorption detection (at 308 nm) and using a multipass optical system. Time-dependent
profile decays were fitted with a 23-step mechanism, but only OH + NO2, OH + HO2, both HO2 and NO2
dissociations, and the atom molecule reactions, O + NO2 and O + C2H4, contributed to the decay profile.
Since all of the reactions except the first two are known with good accuracy, the profiles were fitted by
varying only OH + NO2 and OH + HO2. The new ATcT approach was used to evaluate equilibrium constants
so that back reactions were accurately taken into account. The combined rate constant from the present work
and earlier work by Glaenzer and Troe (GT) is kOH+NO ) 2.25 × 10-11 exp(-3831 K/T) cm3 molecule-1 s-1,
2
which is a factor of 2 lower than the extrapolated direct value from Howard but agrees well with NO + HO2
f OH + NO2 transformed with the updated equilibrium constants. Also, the rate constant for OH + HO2
suitable for combustion modeling applications over the T range (1200-1700 K) is (5 ( 3) × 10-11 cm3
molecule-1 s-1. Finally, simulating previous experimental results of GT using our updated mechanism, we
suggest a constant rate for kHO +NO ) (2.2 ( 0.7) × 10-11 cm3 molecule-1 s-1 over the T range 1350-1760
2
2
K.
1. Introduction
and 9.5 with the mean value being 7.9 × 10-11 cm3 molecule-1
s-1. The temperature dependence was measured in two of the
studies,1,9 and the agreement was excellent over the overlapping
temperature range. There can be no doubt that the NO + HO2
rate constants are accurate from 206 to 1271 K. Hence, the
reported rate constants for the reverse reaction, OH + NO2,
In kinetics studies1 of the reactions, NO + HO2 f OH +
NO2, and its reverse over the respective temperature ranges of
232-1271 K and 452-1115 K, the heat of formation at 298 K
for HO2 radicals was determined to be 2.5 ( 0.6 kcal mol-1
,
implying ∆fH° ) 3.2 ( 0.6 kcal mol-1. This value was
0
are incompatible with the updated ∆fH° for HO2, and this
0
accepted by most workers and was a large change from the
supplies the motivation for the present study.
values recommended in the JANAF compilation (∆fH2°98 ) 0.5
( 2.0 and ∆fH° ) 1.2 ( 2.0, both in kcal mol-1).2 As
0
2. Experimental Section
thoroughly discussed in the preceding article in this volume,3
later experiments and evaluations caused significant fluctuations
in this heat of formation. The preferred values, as assessed with
the Active Thermochemical Table (ATcT) procedure,3 are now
∆fH° ) 2.94 ( 0.06 and ∆fH° ) 3.64 ( 0.06, both again
2.1. Approach. We described earlier a long absorption path
multipass optical system for OH-radical detection in the reflected
shock waves11,12 and used it to measure high-temperature rate
constants.11,13,14 With a path length of 2.798 m, high sensitivity
for OH-radical detection is possible thereby minimizing the
effects of secondary reaction perturbations. In this work, the
same method used previously was employed. OH-radicals were
generated by the H + NO2 reaction13 from controlled H-atom
formation using C2H5I decomposition,15 that is,
298
0
being in kcal mol-1. We noted3 that the 0.44 kcal mol-1 change
from the Howard value would then suggest that either the
measured NO + HO2 rate constant was low by ∼2 or the
measured OH + NO2 rate constant was high by ∼2. Since the
NO + HO2 reaction is of major importance in stratospheric
ozone depletion,4 this rate constant has been measured several
times.5-10 The values at room temperature range between 6.1
C2H5I (+ M) f C2H4 + H + I (+ M)
C2H5I (+ M) f C2H4 + HI (+ M)
H + NO2 f OH + NO
(1)
(2)
(3)
† Part of the special issue “David M. Golden Festschrift”.
* To whom correspondence should be addressed. E-mail: jmichael@
anl.gov.
‡ Chemistry Division, Argonne National Laboratory.
§ Faculty Research Participant, Department of Educational Programs,
Argonne National Laboratory. Permanent address: Department of Chem-
istry, Sonoma State University, Rohnert Park, CA 94928.
| Present address: Guest Scientist, Department of Energy and Technol-
ogy, Brookhaven National Laboratory, Upton, NY 11973.
Reaction 1 is composite since the initially formed C2H5 radicals
instantaneously decompose under the present high-temperature
conditions. If [NO2] is large, then reaction 3 is likewise nearly
10.1021/jp057461x CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/01/2006

Shock Tube Studies for Bimolecular Reactions
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6603
instantaneous. Therefore, the thermal decomposition of C2H5I
controls the rate of formation of OH. If the temperature is high
enough so that the thermal decomposition is fast, then a pulse
of OH at short times is formed, and this property was then used
to obtain the temperature dependence of the absorption cross
section.11 Clearly, if [NO2] is in large enough excess, the effects
of the OH + NO2 reaction should be reflected in the OH-radical
concentration profile, and with this strategy, a decision on the
abovementioned incompatibility between forward and reverse
rates may be possible.
2.2. Instrumentation. The present experiments were per-
formed with the shock tube technique using OH-radical
electronic absorption detection. The method and the apparatus
currently being used have been previously described,16,17 and
only a brief description of the experiment will be presented here.
The shock tube is constructed from 304 stainless steel in three
sections. The first 10.2 cm-o.d. cylindrical section is separated
from the He driver chamber by a 4 mil unscored 1100-H18
aluminum diaphragm. A 0.25 m transition section then connects
the first and third sections. The third section is of rounded corner
(radius, 1.71 cm) square design and is fabricated from flat stock
(3 mm) with a mirror finish. The tube length is 7 m, and the
driver to driven volume ratio is about 3:1.18 Two sets of flat
fused silica windows (3.81 cm) with broadband antireflection
(BB AR) coating for UV light are mounted on the tube across
from one another at a distance of 6 cm from the end plate. The
path length between windows is 8.745 cm. The incident shock
velocity is measured with eight fast pressure transducers (PCB
Piezotronics, Inc., Model 113A21) mounted along the third
portion of the shock tube, and temperature and density in the
reflected shock wave regime are calculated from this velocity
and include corrections for boundary layer perturbations.18-21
Because the tube is long, the shock waves do not attenuate
(giving constant velocities) with the loading pressures used in
this investigation. The tube is routinely pumped between
experiments to <10-8 Torr by an Edwards Vacuum Products
Model CR100P packaged pumping system. A 4094C Nicolet
digital oscilloscope was used to record both the velocity and
absorption signals.
Figure 1. Sample temporal profile of OH absorption: solid line, fit
with the reaction model of Table 1; dashed line, simulation with the
full model but with kOH+NO ) 0. P1 ) 30.58 Torr and Ms ) 2.430. T5
2
) 1461 K, F5 ) 5.588 × 1018 molecules cm-3, [NO2]0 ) 5.672 × 1014
molecules cm-3, and [C2H5I] ) 3.915 × 1013 molecules cm-3
.
concentration so as to not perturb the OH-radical profiles.
Distilled water, evaporated at 1 atm into ultrahigh purity grade
Ar (99.999%) from AGA Gases, was used at ∼25 Torr pressure
in the resonance lamp. Analytical grade C2H5I (99%), from
Aldrich Chemical Co. Inc., and technical grade NO2, from
Matheson Gas Products, were further purified by bulb-to-bulb
distillations with the middle thirds being retained. Test gas
mixtures were accurately prepared from pressure measurements
using a Baratron capacitance manometer and were stored in an
all-glass vacuum line.
3. Results and Discussion
3.1. Kinetics. In this work, the temporal concentration profiles
of OH radicals were measured after the initial [OH] pulse
generated by eqs 1-3. A typical OH profile is shown in Figure
1 where the concentration is seen to slowly decay as time
increases. Uncertainty due to schlieren interference has a
minimal effect on the slow long-time temporal profile. This
long-time behavior was noted previously13 and was attributed
to OH-radical termination by the self-reaction, OH + OH f
H2O + O. However, this reaction has been studied both
directly23 and with O + H2O f OH + OH transformed through
equilibrium constants.16 With the recent reevaluation for the heat
of formation for OH radicals,24,25 the direct and transformed
rate constants now agree with one another, over the present
temperature range, 1237-1554 K, to within <8%, and hence,
this rate constant is accurately known. Including eqs 1-3 and
the self-reaction in an extended mechanism shows that the
decreasing [OH]t behavior is not due to the self-reaction. We
tested other possible causes (e.g., OH + C2H4) eventually
building the 15 species-23-step mechanism shown in Table
1.26-35 With the low [OH] used in the present study, only a
few radical-radical and radical-molecule bimolecular reactions
are important as illustrated in the [OH] sensitivity plot shown
in Figure 2. [OH]t is most sensitive to OH + NO2 with
significant contributions from OH + HO2, both HO2 and NO2
dissociations, and the atom molecule reactions, O + NO2 and
O + C2H4.
The optical configuration consists of an OH resonance
lamp,11,13,22 multipass reflectors, an interference filter at 308
nm, and a photomultiplier tube (1P28) all mounted external to
the shock tube as described previously.11,13 At the entrance to
the multipass cell, the resultant OH resonance radiation was
collimated with a set of lenses and was focused onto the reflector
on the opposite side of the shock tube through two AR coated
windows that were flush mounted to the inside of the shock
tube. The reflectors and windows were obtained from the CVI
Laser Corporation. These reflectors were attached to adjustable
mounts, and the center points of windows and mirrors were all
in a coaxial position. With this new configuration, we were able
to obtain 32 passes, giving a total path length of 2.798 m,
thereby amplifying the measured absorbances. Thus, a substan-
tially better sensitivity (at least a factor of 2) is achieved for
OH-radical detection than in the previous work.13,22 In these
experiments, particularly at the higher pressures, zero time can
only be determined to within ∼ (10 µs due to schlieren
interference.
2.3. Gases. High-purity He (99.995%), used as the driver gas,
was from AGA Gases. Scientific grade Kr (99.999%), the
diluent gas in reactant mixtures, was from Spectra Gases, Inc.
The ∼10 ppm impurities (N2, 2 ppm; O2, 0.5 ppm; Ar, 2 ppm;
CO2, 0.5 ppm; H2, 0.5 ppm; CH4, 0.5 ppm, H2O, 0.5 ppm; Xe,
5 ppm; CF4, 0.5 ppm) are all either inert or in sufficiently low
Before reliable rate constants can be derived from profile fits,
the accuracy of the sensitive secondary processes should be
considered. Earlier values for HO2 dissociation in N2 vary by
∼2 with Tsang and Hampson36 being lower than either the GRI

6604 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Srinivasan et al.
ref
TABLE 1: Mechanism for Determining [OH]t
reaction
rate constanta
C2H5I (+ Kr) f C2H4 + H + I + Kr
C2H5I + Kr f C2H4 + HI + Kr
H + NO2 f OH + NO
H + HI f H2 + I
k1 ) 6.34 × 109 exp(-15894 K/T)
15
15
13
26
k2 ) 0.15 × k1
k3 ) 1.47 × 10-10
k4 ) 9.29 × 10-11 exp(-59 K/T)
H2 + I f H + HI
k5 ) 4.52 × 10-10 exp(-17070 K/T)
k6 ) 7.19 × 10-21 T2.7 exp (917 K/T)
k7 ) 7.48 × 10-20 T2.7 exp(-7323 K/T)
k8 ) 1.62 × 10-10 exp(-7474 K/T)
k9 ) 5.42 × 10-13 T0.375 exp(950 K/T)
k10 ) 8.44 × 10-20 T2.67 exp(-3167 K/T)
k11 ) 3.78 × 10-20 T2.67 exp(-2393 K/T)
k12 ) 3.56 × 10-16 T1.52 exp(-1736 K/T)
k13 ) 1.56 × 10-15 T1.52 exp(-9083 K/T)
k14 ) 6.61 × 10-9 exp(-30189 K/T)
k15 ) 4.21 × 10-12 exp(273 K/T)
26
OH + OH f O + H2O
O + H2O f OH + OH
H + O2 f OH + O
16, 23, 24
16, 23, 24
27
16, 24, 27
16, 24
16, 24
28
16, 24, 28
29
30
OH + O f O2 + H
O + H2 f OH + H
OH + H f H2 + O
OH + H2 f H2O + H
H2O + H f OH + H2
NO2 + Kr f NO + O + Kr
NO2 + O f NO + O2
OH + C2H4 f H2O + H + C2H2
O + C2H4 f H + C2H3O
O + C2H4 f OH + H + C2H2
OH + NO2 f NO + HO2
NO + HO2f OH + NO2
HO2 + Kr f H + O2 + Kr
OH + HO2 f O2 + H2O
I + HO2 f HI + O2
k16 ) 3.35 × 10-11 exp(-2990 K/T)
k17 ) (0.39 exp(-123 K/T)) × (2.25 × 10-17 T1.88 exp(-92 K/T))
k18 ) (1-0.39 exp(-123 K/T)) × (2.25 × 10-17 T1.88 exp(-92 K/T))
k19 ) to be fitted
31
32, 33
32, 33
k20 ) k19 (0.229731 exp(3571 K/T))
k21 ) 7.614 × 10-10 exp(-22520 K/T)
k22 ) to be fitted
34
35
k23 ) 1.47 × 10-11 exp(-1090 K/T)
a All rate constants are in units cm3 molecule-1 s-1
.
Mech37 or Leeds38 databases. The reverse reaction in seven bath
gases including Kr was recently studied in this laboratory.34 In
that work, the relative collision efficiencies were determined,
and new and existing third-order rate constant determinations
were reviewed and fitted with a theoretical model. The values,
from transformations through equilibrium constants, for HO2
dissociation in N2 were 10-20% higher than Tsang and
Hampson, 35-55% lower than GRI Mech, and 70-110% lower
than Leeds. The success of the model for a variety of bath
gases34 leads us to prefer, over the present experimental
temperature range, the transformed rate constant for HO2 + Kr
listed in Table 1.
A similar spread in values exists for the NO2 dissociation
where two competing studies in Ar bath gas have been
reported.29,39 Estimates of the low-pressure limits have been
made with the values from Rohrig et al.29 being ∼1.8-2.6 times
larger than those from Troe.39 Which study is more accurate
can be assessed by considering the direct measurements for the
reverse termolecular reaction, O + NO + Ar, by Yarwood et
al.40 These workers report kO+NO+Ar ) 6.7 × 10-32 T-1.41 cm6
molecule-2 s-1 over the temperature range, 300-1341 K. The
data from the two dissociation studies can be transformed using
equilibrium constants, and as pointed out by Rohrig et al., the
transformed values are within (30% of the Yarwood et al. data.
However, this conclusion was reached using JANAF2 equilib-
rium constants for NO2 S NO + O, but as mentioned earlier
in the previous paper in this issue,3 ATcT now produces new
and more accurate enthalpies of formation for NO2 and NO.
JANAF2 lists ∆fH0°(NO) ) 21.46 ( 0.04 kcal mol-1 (21.58
kcal mol-1 at 298 K) and ∆fH0°(NO2) ) 8.6 ( 0.2 kcal mol-1
(7.9 kcal mol-1 at 298 K), from which ∆EqH0° ) D0(ON-O)
) 71.9 ( 0.2 kcal mol-1 is obtained. As mentioned before,
though the best currently available enthalpies of NO and NO2
(from ATcT) are higher by ∼0.20 kcal mol-1 than those found
in JANAF (and are now both known to be (0.02 kcal mol-1),
the resulting D0(ON-O) ) 71.85 ( 0.03 kcal mol-1 is very
similar and, therefore, the changes from the JANAF equilibrium
constants are fortuitously not large. This is shown in Figure 3
Figure 2. OH-radical sensitivity analysis for the profile shown in
Figure 1 using the full reaction mechanism (Table 1) and the final fitted
values of kOH+NO ) 1.85 × 10-12 and kOH+HO ) 6 × 10-11 both in
Figure 3. Third-order Arrhenius plot for the reaction O + NO + M:
(b) ref 40 (300-1341 K); (2) ref 29 using ref 2; (O) ref 29 using
ATcT (for details see text).
2
2
cm3 molecule-1 s-1. The nine most sensitive reactions are shown in
the inset.

Shock Tube Studies for Bimolecular Reactions
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6605
TABLE 2: High-temperature Rate Data for OH + NO2 and
OH + HO2
F
b
c
c
P1/Torr Msa
5/(1018 cm-3
)
T5/Kb
kOH+NO
kOH+HO
2
2
XC H I ) 7.007 × 10-6
XNO2 ) 1.015 × 10-4
2
30.58 5 2.38
5.527
4.681
3.815
3.433
2.758
5.588
4.981
4.678
3.217
1407 1.35(-12) 7.50(-11)d
1449 1.30(-12) 3.31(-11)
1416 1.75(-12) 9.00(-11)
1502 2.25(-12) 8.00(-11)
1299 1.30(-12) 6.00(-11)
1461 1.85(-12) 6.00(-11)
1237 8.00(-13) 9.00(-11)
1486 2.40(-12) 1.00(-10)
1421 1.50(-12) 4.00(-11)
XNO2 ) 1.360 × 10-4
25.38
20.99
18.26
15.95
30.58
30.13
25.34
17.84
2.42
2.39
2.47
2.27
2.43
2.21
2.45
2.38
XC H I ) 1.046 × 10-5
15.95 5 2.44
2.963
2.736
2.889
2.846
2.025
1.981
1.918
2.062
1482 1.63(-12) 9.50(-11)
1300 1.20(-12) 1.25(-10)
1423 1.35(-12) 9.30(-11)
1393 1.50(-12) 4.00(-11)
1462 2.20(-12) 3.80(-11)
1418 1.35(-12) 9.00(-11)
1357 1.15(-12) 8.00(-11)
1533 1.60(-12) 4.70(-11)
XNO2 ) 5.180 × 10-5
2
15.90
15.91
15.87
10.96
10.91
10.90
10.87
2.27
2.39
2.36
2.42
2.38
2.32
2.48
Figure 4. Arrhenius plot of the data for OH + NO2 from Table 2:
(b) present work (1237-1554 K); (9) ref 43, reanalyzed (see text);
dotted line, ref 1; short dashed line, ref 41 (see text); long dashed line,
ref 41 (see text); dash dots, ref 1 modified using ATcT (see text); thin
solid line, eq 4 in the text.
XC H I ) 1.007 × 10-5
10.92 5 2.47
2.065
2.912
2.091
2.039
1.847
1.930
1.833
1511 2.10(-12) 6.00(-11)
1435 1.73(-12) 3.00(-11)
1554 2.10(-12) 8.50(-11)
1473 1.50(-12) 5.30(-11)
1254 1.40(-12) 6.20(-11)
1340 1.80(-12) 4.00(-11)
1242 9.00(-13) 4.00(-11)
2
15.96
10.89
10.93
10.96
10.99
10.94
2.40
2.51
2.43
2.22
2.30
2.21
a The error in measuring the Mach number, Ms, is typically 0.5-
1.0% at the 1 std dev level giving (1-2% uncertainty in F5 and T5.
b Quantities with the subscript 5 refer to the thermodynamic state of
the gas in the reflected shock region. c Rate constants in units cm3
molecule-1 s-1 d Parentheses denote the power of 10.
.
where the solid circles describing the third-order data of
Yarwood et al.40 are plotted along with the transformed data of
Rohrig et al. using JANAF and ATcT equilibrium constant
values, respectively. Clearly, the old and new values show good
continuity with Yarwood et al. leading us to prefer the Rohrig
et al. low-pressure result for the dissociation (Table 1).
The values for O + NO2 and O + C2H4 listed in Table 1
are, respectively, from a recent study by Estupinan et al.30 and
evaluation by Baulch et al.32 (coupled with the branching ratio
results from Klemm et al.33). The rate constants for many of
the 16 remaining reactions listed in Table 1 have either been
measured or been evaluated in this laboratory.13,15,16,26 In Figure
1, only two important unknown reactions out of the 16 are
sensitive in determining [OH]t. As shown in Figure 2, these
are OH + NO2 f NO + HO2 and OH + HO2 f O2 + H2O
with the former being most important in the 200-600 µs regime
while the latter contributes at later times. Note that the back
reaction, NO + HO2 f OH + NO2, is also taken into account
by using equilibrium constants, again from ATcT. Over the
present temperature range, Keq ) 4.353 exp(-3571 K/T). With
all other rate constants specified, both title reaction rate constants
were varied to determine the best profile fit as shown in Figure
Figure 5. Arrhenius plot for OH + HO2. Inset: (b) present work
(1237-1554 K); Table 2. Full graph: thick solid line, ref 54; dash
dots, ref 60; (*) ref 56; (2) ref 61; dotted line, ref 58; (") ref 57; (0)
ref 59; (×) ref 7; dashed line, ref 48; thin solid line-collision theory
(see text); (b-b) average of present work, (6.7 ( 2.6) × 10-11 cm3
molecule-1 s-1
.
represented as the average, kOH+HO ) (6.7 ( 2.6) × 10-11 cm3
2
molecule-1 s-1
.
The present values for kOH+NO are compared to the discharge
2
flow tube results of Howard1 in Figure 4. In the figure, the
extrapolation of the rate expression given by Howard is seen
to be ∼2 times larger than the present data, as are the theoretical
calculations of Chakraborty et al.41 for the reverse reaction,
transformed with equilibrium constants implied by Howard’s
data. However, if the data1 and theory41 for the back reaction,
NO + HO2, are transformed using equilibrium constants from
the ATcT approach, the agreement with the present data is
excellent. In a recent report,42 a more accurate theoretical
approach has been used to estimate the rate behavior; however,
the results do not agree as well with experiment as the study
by Chakraborty et al. There is one earlier experimental study
1. A profile fit with kOH+NO ) 0 is also plotted, indicating the
2
importance of this reaction. Twenty-four experiments were
performed between 1237 and 1554 K, and the conditions and
resulting rate constant fits for both reactions are given in Table
2. Arrhenius plots are shown in Figures 4 and 5. For OH +
NO2 f NO + HO2, varying rate constants by (20% results in
worse fits while the values for OH + HO2 f O2 + H2O can
vary by (40% before poorer fits result. As seen in the inset of
Figure 5, there is little evidence for temperature dependence
for the OH + HO2 reaction and the present results can be
on kOH+NO at high temperature by Glaenzer and Troe43 (GT).
2

6606 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Srinivasan et al.
These workers monitored HO2 radicals in quite concentrated
mixtures of HNO3 and NO2 in Ar bath gas, and they measured
absolute [HO2] at half its maximum value and the associated
not accurate enough to contribute to this question. Considering
the difficulty in calculating ab initio surfaces as shown in
previous work,49-51,54,55 it may be difficult to theoretically justify
such behavior using available electronic structure and dynamical
methods. In past studies, we have employed a collision theory
method to estimate bimolecular rate constants with no barri-
ers,62,63 and these are generally ∼60% of better calculations that
use accurate potential energy surfaces and flexible transition
state theory.64 The thin line in Figure 5 is a calculation assuming
the triplet pathway for OH + HO2 using collision theory, and
this would be a rough upper estimate if no barrier exists for the
reaction. The point to this illustration is that even if the narrow
minimum exists at 1000 K, one would expect the bimolecular
rate constant at high-T to level off at ∼1 × 10-10 cm3
molecule-1 s-1. Hence, we conclude that the best available
experimental value for OH + HO2 in combustion modeling
applications with 1200 e T e 1700 K is (5 ( 3) × 10-11 cm3
decay time to reach this value. With assumptions for kOH+HO
2
and kHO +NO /kNO+HO , they could estimate values for kOH+NO
2
2
2
2
to within ∼30-40%.
Because the thermochemistry has changed appreciably since
the work of GT, we have reanalyzed their results using the
updated mechanism of Table 1 with the most important HO2
destruction reaction, HO2 + NO2, included. Eight of their
experiments were simulated by mutually varying both kOH+NO
2
and kHO +NO with a constant value of kOH+HO ) 5 × 10-11
2
2
2
cm3 molecule-1 s-1 over their T range. This kOH+HO value is a
2
compromise between the present value and previous high-T
measurements44-48 that give values ranging from ∼1.8-10 ×
10-11 cm3 molecule-1 s-1. The reanalyzed values obtained from
the simulations for kOH+NO are plotted as solid squares in Figure
2
molecule-1 s-1
.
4 and are in excellent agreement with the present data.
Recognizing that kOH+HO is fairly uncertain, we performed a
2
4. Conclusions
sensitivity analysis on this GT data43 which shows that a (40%
change in OH + HO2 results in (20% and (30% uncertainties
in kOH+NO and kHO +NO , respectively.
On the basis of the ATcT approach, in the present work, we
have used the best currently available values for the enthalpy
of formation of HO2, ∆fH0°(HO2) ) 3.64 ( 0.06 kcal mol-1
(∆fH298°(HO2) ) 2.94 ( 0.06 kcal mol-1), and new ATcT
values for several other species, such as NO, NO2, and OH.
New kinetics measurements are reported confirming that the
previous kinetics rate for OH + NO2 f NO + HO2 was too
2
2
2
If the present results are combined with the reanalyzed values
from GT, then the rate constant for OH + NO2 follows the
Arrhenius expression
kOH+NO
)
2
high by a factor of ∼2. The new rate constant is kOH+NO
)
2
2.25 × 10-11 exp(-3831 K/T) cm3 molecule-1 s-1 (4)
2.25 × 10-11 exp(-3831 K/T) cm3 molecule-1 s-1 for the T
range 1200-1700 K. We also conclude that the best available
experimental value for OH + HO2 in combustion modeling
applications within 1200 e T e 1700 K is (5 ( 3) × 10-11
cm3 molecule-1 s-1. Simulating the experimental results of GT,
we suggest that kHO +NO over the T range 1350-1760 K is
for the temperature range, 1200-1700 K. The derived rate
constant for HO2 + NO2 over the GT T range from the
simulations is, within experimental error, constant at kHO +NO
2
2
) (2.2 ( 0.7) × 10-12 cm3 molecule-1 s-1
.
2
2
constant at (2.2 ( 0.7) × 10-11 cm3 molecule-1 s-1
.
There are no theoretical investigations for kHO +NO but several
2
2
for kOH+HO2.
49-55 Earlier determinations for HO2 + NO2 show
Acknowledgment. This work, together with the underlying
thermochemical development of the ATcT approach, was
supported by the U.S. Department of Energy, Division of
Chemical Sciences, Geosciences, and Biosciences of the Office
of Basic Energy Sciences, under Contract No. W-31-109-ENG-
38. Underpinning computer-science and data management
aspects of the ATcT development were supported by the U.S.
Department of Energy, Division of Mathematical, Information,
and Computational Science of the Office of Advanced Scientific
Computing Research, under Contract No. W-31-109-ENG-38
(Argonne) as part of the multi-institutional Collaboratory for
Multi-Scale Chemical Science (CMCS),65 which is a project
within the National Collaboratories Program of the U.S.
Department of Energy. Portions of the research described are
related to the effort of a Task Group of the International Union
of Pure and Applied Chemistry (2003-024-1-100).66
that the reactivity at room temperature is strongly dominated
by the third-order recombination reaction giving the stabilized
product HO2NO2 and not O2 + HNO2. It is likely that the
bimolecular products, first noted by GT, arise from competition,
between forward and backward dissociations of vibrationally
hot HO2NO2*, and stabilization on a singlet potential energy
surface. The most thorough ab initio electronic structure
calculations on OH + HO2 were carried out by Gonzalez et al.
on both the singlet54 and triplet surfaces.55 They concluded that
the singlet pathway through HOOOH was unimportant because
the transition state was 15.2 kcal mol-1 higher lying than
reactants. They identified tight transition states on the triplet
surface corresponding to an abstraction pathway; however, the
implied rate constants were much too low using transition state
theory. They then calculated rate constants using a vibrationally/
rotationally adiabatic capture dipole-dipole model, but these
values had to be scaled by 0.61 to make contact with existing
low-temperature values7,56-61 (ranging from 5.2 to 11.0 × 10-11
cm3 molecule-1 s-1). These experimental and theoretical values
are plotted in Figure 5 along with the present determination.
There are four studies at high temperature by Troe and
co-workers45-48 with the last being the preferred determination48
(shown as the dashed line in Figure 5). These data imply a rapid
increase in the rate constant, between ∼1.2 and 7.0 × 10-11
cm3 molecule-1 s-1, for a small temperature increase, ∼970-
1220 K. This and earlier data47 suggested a narrow minimum
in the rate constant over a small T range with the minimum
being at ∼1000 K. There are no data in the intermediate T range,
420-950 K, to confirm this behavior, and the present data are
References and Notes
(1) Howard, C. J. J. Am. Chem. Soc. 1980, 102, 6937.
(2) Chase, M. W., Jr.; Davies, C. A.; Downey, J. R., Jr.; Frurip, D. J.;
McDonald, R. A.; Syverud, A. N. J. Phys. Chem. Ref. Data 1985, 14, Suppl.
1. Chase, M. W., Jr. J. Phys. Chem. Ref. Data 1998, Monograph 9.
(3) Ruscic, B.; Pinzon, R. E.; Morton, M. E.; Srinivasan, N. K.; Su,
M.-C.; Sutherland, J. W.; Michael, J. V. J. Phys. Chem. A 2006, 110, 6592.
(4) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling; Evaluation No. 12 JPL Publication 97-4; Jet Propulsion Labora-
tory: Pasadena, CA, 1997.
(5) Hack, W.; Preuss, A. W.; Temps, F.; Wagner, H. Gg.; Hoyermann,
K. Int. J. Chem. Kinet. 1980, 12, 851.
(6) Thrush, B. A.; Wilkinson, J. P. T. Chem. Phys. Lett. 1981, 81, 1.

Shock Tube Studies for Bimolecular Reactions
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6607
(7) Rozenshtein, V. B.; Gershenzon, Yu. M.; Il’in, S. D.; Kishkovitch,
O. P. Chem. Phys. Lett. 1984, 112, 473.
(34) Michael, J. V.; Su, M.-C.; Sutherland, J. W.; Carroll, J. J.; Wagner,
A. F. J. Phys. Chem. A 2002, 106, 5297.
(8) Jemi-Alade, A. A.; Thrush, B. A. J. Chem. Soc., Faraday Trans.
1990, 86, 3355.
(9) Seeley, J. V.; Meads, R. F.; Elrod, M. J.; Molina, M. J. J. Phys.
Chem. 1996, 100, 4026.
(10) Bohn, B.; Zetzsch, C. J. Chem. Soc., Faraday Trans. 1998, 94,
1203.
(11) Srinivasan, N. K.; Su, M.-C.; Sutherland, J. W.; Michael, J. V. J.
Phys. Chem. A 2005, 109, 1857.
(35) Jenkin, M. E.; Cox, R. A.; Mellouki, A.; Le Bras, G.; Poulet, G. J.
Phys. Chem. 1990, 94, 2927.
(36) Tsang, W.; Hampson, R. F. J. Phys. Chem. Ref. Data 1986, 15,
1087.
(37) http://www.me.berkeley.edu/gri_mech/.
(38) http://chem.leeds.ac.uk/combustion/combustion.html/.
(39) Troe, J. Ber. Bunsen-Ges. Phys. Chem. 1969, 73, 144.
(40) Yarwood, G.; Sutherland, J. W.; Wickramaaratchi, M. A.; Klemm,
R. B. J. Phys. Chem. 1991, 95, 8771.
(41) Chakraborty, D.; Park, J.; Lin, M. C. Chem. Phys. 1998, 231, 39.
(42) Zhu, R. S.; Lin, M. C. J. Chem. Phys. 2003, 119, 10667.
(43) Glaenzer, K.; Troe, J. Ber. Bunsen-Ges. Phys. Chem. 1975, 79,
465.
(12) Su, M.-C.; Kumaran, S. S.; Lim, K. P.; Michael, J. V. ReV. Sci.
Instrum. 1995, 66, 4649.
(13) Su, M.-C.; Kumaran, S. S.; Lim, K. P.; Michael, J. V.; Wagner, A.
F.; Harding, L. B.; Fang, D.-C. J. Phys. Chem. A 2002, 106, 8261.
(14) Srinivasan, N. K.; Su, M.-C.; Sutherland, J. W.; Michael, J. V. J.
Phys. Chem. A 2005, 109, 7902.
(15) Kumaran, S. S.; Su, M.-C.; Lim, K. P.; Michael, J. V. Proc.
Combust. Inst. 1996, 26, 605.
(44) Goodings, J. N.; Hayhurst, A. N. J. Chem. Soc., Faraday Trans. 2
1988, 84, 745.
(16) Michael, J. V. Prog. Energy Combust. Sci. 1992, 18, 327.
(17) Michael, J. V. In AdVances in Chemical Kinetics and Dynamics;
Barker, J. R., Ed.; JAI Press: Greenwich, CT, 1992; Vol. I, pp 47-112,
for original references.
(18) Michael, J. V.; Kumaran, S. S.; Su, M.-C. J. Phys. Chem. A 2000,
104, 980.
(19) Michael, J. V.; Sutherland, J. W. Int. J. Chem. Kinet. 1986, 18,
409.
(20) Michael, J. V. J. Chem. Phys. 1989, 90, 189.
(21) Michael, J. V.; Fisher, J. R. In SeVenteenth International Symposium
on Shock WaVes and Shock Tubes; Kim, Y. W., Ed.; AIP Conference
Proceedings 208; American Institute of Physics: New York, 1990; pp 210-
215.
(45) Hippler, H.; Troe, J. J. Chem. Phys. 1990, 93, 1755.
(46) Hippler, H.; Troe, J. Chem. Phys. Lett. 1992, 192, 333.
(47) Hippler, H.; Neunaber, H.; Troe, J. J. Chem. Phys. 1995, 103, 3510.
(48) Kappel, Ch.; Luther, K.; Troe, J. Phys. Chem. Chem. Phys. 2002,
4, 4392.
(49) Cremer, D. J. Chem. Phys. 1978, 69, 4456.
(50) Jackels, C. F.; Phillips, D. H. J. Chem. Phys. 1986, 84, 5013.
(51) Toohey, D. W.; Anderson, J. G. J. Phys. Chem. 1989, 93, 1094.
(52) Mozurkewich, M. J. Phys. Chem. 1986, 90, 2216.
(53) Phillips, L. F. J. Phys. Chem. 1990, 94, 7482.
(54) Gonzalez, C.; Theisen, J.; Zhu, L.; Schlegel, H. B.; Hase, W. L.;
Kaiser, E. W. J. Phys. Chem. 1991, 95, 6784.
(55) Gonzalez, C.; Theisen, J.; Schlegel, H. B.; Hase, W. L.; Kaiser, E.
W. J. Phys. Chem. 1992, 96, 1767.
(56) Cox, R. A.; Burrows, J. P.; Wallington, R. J. Chem. Phys. Lett.
1981, 84, 217.
(22) Krasnoperov, L. N.; Michael, J. V. J. Phys. Chem. A 2004, 108,
5643.
(23) Wooldridge, M. S.; Hanson, R. K.; Bowman, C. T. Int. J. Chem.
Kinet. 1994, 26, 389.
(24) Ruscic, B.; Wagner, A. F.; Harding, L. B.; Asher, R. L.; Feller,
D.; Dixon, D. A.; Peterson, K. A.; Song, Y.; Qian, X.; Ng, C. Y.; Liu, J.;
Chen, W.; Schwenke, D. W. J. Phys. Chem. A 2002, 106, 2727.
(25) Herbon, J. T.; Hanson, R. K.; Golden, D. M.; Bowman, C. T. Proc.
Combust. Inst. 2002, 29, 1201.
(26) Michael, J. V.; Kumaran, S. S.; Su, M.-C.; Lim, K. P. Chem. Phys.
Lett. 2000, 319, 99.
(27) Du, H.; Hessler, J. P. J. Chem. Phys. 1992, 96, 1077.
(28) Oldenborg, R. C.; Loge, G. W.; Harradine, D. M.; Winn, K. R. J.
Phys. Chem. 1992, 96, 8426.
(29) Rohrig, M.; Petersen, E. L.; Davidson, D. F.; Hanson, R. K. Int. J.
Chem. Kinet. 1997, 29, 483.
(57) Temps, F.; Wagner, H. Gg. Ber. Bunsen-Ges. Phys. Chem. 1982,
86, 119.
(58) Sridharan, U. C.; Qiu, L. X.; Kaufman, F. J. Phys. Chem. 1984,
88, 1281.
(59) Dransfeld, P.; Wagner, H. Gg. Z. Naturforsch. A: Phys. Sci. 1987,
42, 471.
(60) Keyser, L. F. J. Phys. Chem. 1988, 92, 1193.
(61) Schwab, J. J.; Brune, W. H.; Anderson, J. G. J. Phys. Chem. 1989,
93, 1030.
(62) Kumaran, S. S.; Su, M.-C.; Lim, K. P.; Michael, J. V.; Wagner, A.
F.; Harding, L. B. J. Phys. Chem. 1996, 100, 7541.
(63) Sutherland, J. W.; Su, M.-C.; Michael, J. V. Int. J. Chem. Kinet.
2001, 33, 669.
(64) VARIFLEX, version 1.0 (http://chemistry.anl.gov/chem-dyn/Vari-
Flex), VARIFLEX is a freeware program package for calculating gas-phase
reaction rates written by Klippenstein, S. J.; Wagner, A. F.; Robertson, S.
H.; Dunbar, R. C.; Wardlaw, D. M. in 1999.
(30) Estupinan, E. G.; Nicovich, J. M.; Wine, P. H. J. Phys. Chem. A
2001, 105, 9697.
(31) Tully, F. P. Chem. Phys. Lett. 1988, 143, 510.
(32) Baulch, D. L.; Cobos, C. J.; Cox, R. A.; Frank, P.; Hayman, G.;
Just, Th.; Kerr, J. A.; Murrells, T.; Pilling, M. J.; Troe, J.; Walker, R. W.;
Warnatz, J. J. Phys. Chem. Ref. Data 1994, 23, 847.
(33) Klemm, R. B.; Sutherland, J. W.; Wickramaaratchi, M. A.;
Yarwood, G. J. Phys. Chem. 1990, 94, 3354.
(65) http://cmcs.org
(66) http://www.iupac.org/projects/2003/2003-024-1-100.html