High-temperature shock tube study of the reactions CH3 + OH → products and CH3OH + Ar → products

Article abstract of DOI:10.1002/kin.20334

The reaction between methyl and hydroxyl radicals has been studied in reflected shock wave experiments using narrow-linewidth OH laser absorption. OH radicals were generated by the rapid thermal decomposition of tert-butyl hydroperoxide. Two different species were used as CH₃ radical precursors, azomethane and methyl iodide. The overall rate coefficient of the CH₃ + OH reaction was determined in the temperature range 1081-1426 K under conditions of chemical isolation. The experimental data are in good agreement with a recent theoretical study of the reaction. The decomposition of methanol to methyl and OH radicals was also investigated behind reflected shock waves. The current measurements are in good agreement with a recent experimental study and a master equation simulation.

Full text of DOI:10.1002/kin.20334

High-Temperature Shock
Tube Study of the Reactions
CH3 + OH → Products and
CH3OH + Ar → Products
VENKATESH VASUDEVAN, ROBERT D. COOK, RONALD K. HANSON, CRAIG T. BOWMAN,
DAVID M. GOLDEN
High Temperature Gasdynamics Laboratory, Mechanical Engineering Department, Stanford University,
Stanford, CA 94305
Received 9 January 2008; revised 28 February 2008; accepted 29 February 2008
DOI 10.1002/kin.20334
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: The reaction between methyl and hydroxyl radicals has been studied in reflected
shock wave experiments using narrow-linewidth OH laser absorption. OH radicals were gen-
erated by the rapid thermal decomposition of tert-butyl hydroperoxide. Two different species
were used as CH3 radical precursors, azomethane and methyl iodide. The overall rate coeffi-
cient of the CH3 + OH reaction was determined in the temperature range 1081–1426 K under
conditions of chemical isolation. The experimental data are in good agreement with a recent
theoretical study of the reaction. The decomposition of methanol to methyl and OH radicals
was also investigated behind reflected shock waves. The current measurements are in good
agreement with a recent experimental study and a master equation simulation. ꢀ 2008 Wiley
C
Periodicals, Inc. Int J Chem Kinet 40: 488–495, 2008
actants, or react via one of several chemical activation
channels. The following product pathways have been
identified using high-level theoretical methods [1,3,4]:
INTRODUCTION
The reaction between CH3 and OH plays an impor-
tant role in the modeling of hydrocarbon flames and in
the oxidation of methanol [1]. This reaction was also
a secondary reaction encountered in a recent study of
the CH3 + O2 reaction [2]. The CH3 + OH reaction
involves the formation of a vibrationally excited inter-
mediate, CH3OH∗, which may stabilize via collisions
to form methanol [CH3OH], dissociate to form the re-
CH3 + OH → CH3OH
(1a)
(1b)
(1c)
(1d)
(1e)
(1f)
(1g)
1
→ CH2 + H2O
→ CH3O + H
→ CH2OH + H
→ CH2O + H2
Correspondence to: Venkatesh Vasudevan, Corporate Strategic
Research, ExxonMobil Research and Engineering Company,
Annandale, NJ 08801; e-mail: venkatesh.vasudevan@exxonmobil.
com.
→ cis-HCOH + H2
→ trans-HCOH + H2
Contract grant sponsor: National Science Foundation.
Contract grant number: 0308700.
2008 Wiley Periodicals, Inc.
c
ꢀ
CH3 and OH can also react directly via the abstraction

CH3 + OH AND CH3OH + Ar REACTION RATE COEFFICIENTS
489
reaction,
This variation in the reported high-temperature data
is the motivation for the current study of methanol
decomposition.
In the present work, the overall rate coefficient of
reaction (1) and the rate coefficient of reaction (2a)
have been measured behind reflected shock waves us-
ing narrow linewidth ring-dye laser absorption of OH
at 306.7 nm. The results have been compared to previ-
ous experiments and theory.
3
CH3 + OH → CH2 + H2O
(1h)
Although there have been several previous studies
[1 and references cited therein] of reaction (1) at low-
to-moderate temperatures (<1000 K), there have been
only a few direct high-temperature measurements of
this reaction system [5–7]. Bott and Cohen [5] studied
reaction (1) at 1200 K and 1 atm in a shock tube using
UV absorption measurements of OH resonance radia-
tion at 309 nm and reported an overall rate coefficient of
Experimental Setup
1.1 × 1013 cm3 mol−1
s
−1. Krasnoperov and Michael
Experiments were carried out in the reflected shock
region of a high-purity, stainless steel, helium-driven
shock tube. Further details of the shock tube setup are
available elsewhere [2,11,12]. Incident shock veloc-
ity measurements were made using five PZT pressure
transducers and four programmable timer counters, and
linearly extrapolated to the endwall. Reflected shock
conditions were calculated from the measured shock
velocity using ideal shock relations.
A commercially available solution of 70% tert-butyl
hydroperoxide (TBHP) in water from Sigma Aldrich
(St. Louis, MO, USA) was used as the OH precursor
in the current experiments. Kinetic simulations show
that water vapor in the initial reactant mixture has
little or no effect on the rate coefficient measurements.
Research-grade argon (99.999%) was supplied by
Praxair, Inc. (Danbury, CT, USA). Methanol (>99%)
and methyl iodide (>99.5%) were obtained from
Sigma Aldrich. Azomethane was prepared and
purified using standard methods that are described
in the literature [13–15]. Mixtures were prepared in
a mixing chamber equipped with a magnetic stirrer
assembly. Preshock mixture samples were analyzed in
a gas chromatograph using a flame ionization detector
to check the possible decomposition of TBHP in the
gas phase in the mixing chamber. These measurements
show that less than 0.30 ppm TBHP decomposed to
form acetone [11], and this has no discernible effect
on our rate coefficient determination.
[6] used multipass absorption of OH resonance radia-
tion at 308 nm to measure k1 behind reflected shock
waves at 800–1200 K and 1800–2400 K for pressures
between 50 and 940 Torr. The two-parameter least-
squares fit recommended by Krasnoperov and Michael
yields a rate coefficient that is about a factor of 2 larger
than the Bott and Cohen measurement at 1200 K. Most
recently, reaction (1) was studied by Srinivasan et al.
[7] using a similar multipass OH absorption detection
system, but with increased path length. Srinivasan et al.
report an overall rate of 7.8 × 1012 cm3 mol−1 s−1 in
the 1000–1200 K temperature range, ∼40% lower than
Bott and Cohen and approximately a factor of 3 lower
than Krasnoperov and Michael. It is evident that there
is much uncertainty in k1 at elevated temperatures of
interest in combustion. In this paper, we report direct
measurements of the overall rate coefficient of reaction
(1) in the 1081–1426 K temperature range.
Methanol is an important intermediate species in
hydrocarbon flames, being produced via reaction (1a),
and it is of interest as an alternative fuel [3,8]. The
reactions of methanol comprise an important subset
of detailed hydrocarbon combustion mechanisms. The
thermal decomposition of methanol has been studied
previously [9,10]. Seven decomposition pathways have
been identified using theory [3,4], but at combustion-
relevant conditions the dominant channels are
OH radicals were monitored using a narrow
linewidth ring-dye laser tuned to the center of the R1
(5) absorption line in the OH A-X (0,0) band near
306.7 nm. The diagnostic system has been described
in detail elsewhere [11,12]. A 532-nm solid-state laser
was used to pump a tunable ring-dye laser cavity gener-
ating visible light at ∼614 nm. The visible light beam
was intracavity frequency-doubled in a temperature-
tuned doubling crystal producing ∼1.5 mW of UV
light at ∼306.7 nm. Quantitative OH concentration
profiles were generated from the raw traces of frac-
tional transmission using Beer’s law. The OH diagnos-
tic has a minimum absorption detectivity of ∼0.1%,
CH3OH + M → CH3 + OH + M
(2a)
(2b)
1
→ CH2 + H2O + M
The two most recent measurements of the methanol de-
composition reaction were performed by Krasnoperov
and Michael [6] and Srinivasan et al. [7]. In both stud-
ies, OH profiles during methanol decomposition were
recorded using multipass absorption spectrometry. The
reported overall rates in [6] are about a factor of 2 larger
than in [7] in the temperature range 1700–2300 K,
whereas the recent Baulch et al. evaluation [9] rec-
ommends rates that are 1.6–2.5 times higher than [7].
International Journal of Chemical Kinetics DOI 10.1002/kin

490
VASUDEVAN ET AL.
and is capable of ppm-level OH detection with mi-
crosecond timescale resolution.
Rate coefficients for reaction (1) were inferred by
matching measured and modeled OH time histories
behind the reflected shock wave. The reaction mecha-
nism reported by Srinivasan et al. [7] with the following
modifications was used in the modeling. Updated rate
coefficients from GRI-Mech 3.0 [18] were used for
methyl radical recombination
CH3 + OH Kinetics
The overall rate for reaction (1), CH3 + OH → Prod-
ucts, was measured behind reflected shock waves us-
ing OH laser absorption at temperatures between 1081
and 1426 K. Known and reproducible amounts of CH3
radicals were generated by shock-heating azomethane
(C2H6N2) or methyl iodide (CH3I), whereas OH radi-
cals were generatedusingTBHP[(CH3)3 CO OH] as
the precursor species. Azomethane, upon shock heat-
ing, nearly instantaneously decomposes to form two
CH3 radicals and N2 via reaction (3), while methyl io-
dide decomposes to form a methyl radical and I, via
reaction (4),
2CH3(+M) → C2H6(+M)
and the OH self-reaction
(7)
2OH → H2O + O
(8)
The updated rate coefficients for reactions (7) and (8)
are in good agreement with the recent recommenda-
tions of Baulch et al. [9].
In addition, reactions describing azomethane de-
composition [19] and TBHP chemistry [11,16,17] were
added to the mechanism. Table I presents these and
other reactions that are important in the CH3 + OH
experiments. The recently updated heat of formation
of OH [20–22] was used in the modeling.
C2H6N2 → CH3 + CH3 + N2
CH3I → CH3 + I
(3)
(4)
The decomposition of TBHP yields an OH radical, a
CH3 radical, and acetone [11,16].
(CH3)3 CO OH → (CH3)3CO + OH
(CH3)3CO → (CH3)2CO + CH3
(5)
(6)
Azomethane Experiments
With azomethane as the methyl radical precursor, ex-
periments were carried out over the 1105–1426 K
temperature range. The nominal pressure for most ex-
periments was ∼1.65 atm. One experiment was con-
ducted at a pressure of ∼5 atm to investigate the
possible dependence of the rate parameter on pres-
sure. Figure 1a presents a typical OH concentration
time history obtained upon shock heating a mixture of
21.5 ppm TBHP, 75 ppm azomethane, and argon, and
Fig. 1b is a sensitivity plot for this experiment. At
1245 K and 1.68 atm, an overall rate coefficient of
1.70 × 1013 cm3 mol−1 s−1 for reaction (1) (solid line)
leads to excellent agreement between model and ex-
periment. The two dashed lines show the effect of a
factor of 2 variation in k1 to indicate the sensitivity of
the OH trace to the value of k1. In the kinetic model-
ing, the products of reaction (1) were assumed to be
1CH2 and H2O, reaction (1b). However, the choice of
products has no discernible effect on our overall rate
coefficient determination. For example, if in the kinetic
Nominal mixtures of TBHP and azomethane or methyl
iodide in argon were prepared. Loss of TBHP due
to wall adsorption and condensation effects was ob-
served. To understand the origin of these losses, exper-
iments were performed with and without shock tube
wall passivation. The shock tube walls were passivated
by filling the driven section of the shock tube to the
desired fill pressure with the reactant mixture. After a
period of time, the shock tube was pumped out using a
roughing pump. In the ensuing experiment, carried out
immediately after passivation, the mixture was shock-
heated within 30–60 s of filling the shock tube. Peak
OH yields, with and without passivation, were found
to differ by less than 5% at comparable reflected shock
conditions. This suggests that the loss of TBHP due to
wall adsorption and condensation effects in the shock
tube is small, and the observed loss occurs primar-
ily in the mixing chamber. Since GC analyses of the
preshock reactant mixtures show that TBHP does not
decompose in the gas phase in the mixing tank, the
measurement of OH behind the reflected shock is an
accurate indicator of the concentration of TBHP in the
shock tube. In the kinetic modeling, an initial TBHP
concentration that resulted in the measured peak OH
yield was used; the approach is described in greater
detail elsewhere [11].
1
model the product pathways are taken to be CH2 +
H2O, reaction (1b), and CH3OH, reaction (1a), with
branching fractions of 0.50 each, the inferred overall
rate coefficient remains unchanged.
The OH sensitivity analysis presented in Fig. 1b
shows that CH3 + OH → products is the most sensi-
tive reaction over the entire experimental time frame.
Excellent isolation of the target reaction was achieved
International Journal of Chemical Kinetics DOI 10.1002/kin

CH3 + OH AND CH3OH + Ar REACTION RATE COEFFICIENTS
491
Table I Rate Coefficients of Reactions Important in CH3 + OH Experimentsa
Rate Coefficient (cm3 mol−1 s1)
Reaction
A
n
E (cal mol−1
)
Reference
CH3 + OH → Products
See text
2.50E+15
This work
[11]
[16]
[18]
[7]
(CH3)3 CO OH → (CH3)3CO + OHb
0.0
0.0
43,017
15,300
b
(CH3)3CO → (CH3)2CO + CH3
1.30E+14
2CH3 (+M) → C2H6 (+M)
Pressure dependent
1.614E+06
2.220E+39
2.951E+13
3.57E+04
OH + C2H6 → C2H4 + H + H2O
2.224
−7.99
0.0
2.4
0.0
741
51,505
4,564
−2,110
40,870
−888
18.09
b
C2H6N2 → CH3 + CH3 + N2
[19]
[12]
[18]
[7]
[7]
[7]
CH3COCH3 + OH → CH3COCH2 + H2O
OH + OH → O + H2O
CH3I + Ar → CH3 + I + Ar
CH3I + OH → CH2I + H2O
3CH2 + OH → CH2O + H
3CH2 + CH3 → C2H4 + H
4.842E+15
1.64E+00
3.97
6.685E+13
2.282E+14
0.01662
−0.1316
16.28
[7]
a For complete mechanism, see [7].
b Rate coefficient units: s−1
because of the low OH concentrations (see Fig. 1a,
vertical axis) used in the experiments. However, there
is slight interference from the following secondary
reactions:
30
25
20
Experiment
Model best-fit, k1
2k1, k1/2
15
10
2CH3(+M) → C2H6(+M)
OH + C2H6 → C2H5 + H2O
(7)
(9)
5
0
CH3COCH3 + OH → CH3COCH2 + H2 (10)
3CH2 + OH → CH2O + H
(11)
(a)
0
50
100
150
200
The effect of the uncertainty in the rate coefficients
of reactions (7) and (9)–(11) was taken into account
in a detailed and systematic error analysis. Various
uncertainty categories were considered. These include
uncertainty in (a) temperature, (b) secondary chem-
istry, (c) time-zero, (d) fitting the modeled trace to
experiment, (e) wavemeter reading, and (f) absorption
coefficient. An uncertainty of ∼ 25% is estimated for
k1 at 1245 K and 1.68 atm.
Time ( s)
CH3+OH
C2H6+OH
2CH3(+M)
1CH2+H2O
C2H4+H2O+H
C2H6(+ M)
(b)
1.0
0.5
CH3COCH3+OH
CH3COCH2+H2O
3CH2+OH
CH2O+H
0.0
−0.5
−1.0
−1.5
−2.0
Methyl Iodide Experiments
Experiments were also conducted using methyl iodide
as a CH3 source for the temperature range 1081–
1425 K and a nominal pressure of ∼1.3 atm. The
decomposition of CH3I is not as rapid as that of
azomethane. Therefore, to reduce secondary interfer-
ence from the reaction between CH3I and OH, the
concentration of CH3I was limited to ∼20 ppm in
the initial reaction mixture. An example experiment at
1263 K and 1.28 atm is shown in Fig. 2. The inferred
−25
0
25
Time ( s)
50
75
Figure 1 Example CH3 + OH experiment with azomethane
as the CH3 precursor: (a) OH concentration time history;
solid black line: best-fit k , dashed lines: factor of two vari-
ation in k . (b) Early time OH sensitivity; 21.5 ppm TBHP,
75 ppm azomethane, balance Ar; Initial reflected shock con-
ditions: 1245 K, 1.68 atm; S = (dXOH/dki)(ki/XOH), where
ki is the rate coefficient for reaction i.
1
1
overall rate is 2.2 × 1013 cm3 mol−1
s
−1; the dashed
lines show the effect of a factor of two variation in k1.
Comparison of Figs. 1a and 2a shows that the isolation
International Journal of Chemical Kinetics DOI 10.1002/kin

492
VASUDEVAN ET AL.
Table II CH3 + OH → Products: Rate Coefficient Data
Experiment
Model best-fit, k1
2k1, k1/2
20
15
10
5
T5 (K)
P5 (atm)
k (cm3 mol−1 s−1
)
1
TBHP, 75 ppm azomethane, balance Ar
1192
1245
1284
1348
1105
1143
1211
1268
1397
1238
1426
1.67
1.68
1.63
1.57
1.72
1.72
1.65
1.65
1.52
1.69
4.95
1.80 × 1013
1.70 × 1013
1.60 × 1013
1.75 × 1013
2.70 × 1013
2.35 × 1013
1.75 × 1013
1.70 × 1013
1.80 × 1013
1.75 × 1013
1.50 × 1013
Fit early time
OH to k1
0
(a)
−50 −25
0
25 50 75 100 125 150 175
Time (µs)
TBHP, 20 ppm methyl iodide, balance Ar
0.1
1263
1147
1081
1339
1425
1.28
1.33
1.42
1.25
1.21
2.20 × 1013
2.80 × 1013
2.90 × 1013
2.50 × 1013
2.60 × 1013
0.0
−0.1
−0.2
−0.3
−0.4
−0.5
CH3+OH
3CH2+OH
CH3COCH3+OH
1CH2+H2O
CH2O+H
CH3COCH2+H2O
C2H4+H
CH3+I+AR
3CH2+CH3
CH3I+AR
OH+CH3I
2CH3(+M)
2OH
CH2I+H2O
C2H6(+M)
O+H2O
Hence, rate coefficient data inferred from the
azomethane precursor experiments are considered
more reliable. Table II summarizes the current mea-
surements of the rate coefficient of reaction (1).
(b)
75
−25
0
25
50
Time (µs)
Figure 2 Example CH3 + OH experiment with methyl
iodide as the CH3 precursor: (a) OH concentration time his-
Methanol Decomposition Kinetics
tory; solid black line: best-fit k , dashed lines: factor of two
1
variation in k ; k is inferred from the early time OH pro-
1
1
The decomposition of methanol was investigated in
the 1904–2298 K temperature range at a nominal re-
flected shock pressure of ∼1.3 atm. Dilute mixtures
of methanol (∼50 ppm) in argon were shock-heated,
and OH radicals were detected behind reflected shock
waves using ring-dye laser absorption. The measured
OH profiles were fitted to detailed kinetic simulations;
the GRI-Mech 3.0 mechanism [18] and the recently
revised heat of formation for OH [20–22] were used in
the modeling.
An example OH concentration profile is shown in
Fig. 3a. At early times (t < 50 μs), the decomposition
of methanol to CH3 and OH shows dominant sensitiv-
ity; see Fig. 3b. The rate coefficient of reaction (2a) was
adjusted in the mechanism until the modeled profile
matches experiment. At 2151 K and 1.23 atm, a rate co-
efficient of 4.94 × 109 cm3 mol−1 s−1 was inferred. The
dashed lines show the effect of a factor of two variation
in the decomposition rate. In the kinetic modeling, the
decomposition of methanol was assumed to proceed
via reactions (2a) and (2b). The theoretical modeling
of Jasper et al. [3] indicates that this is a reasonable as-
sumption for the experimental conditions of the present
work. The rate coefficient expression recommended by
file (t < ∼75 μs), at later times there is increased inter-
ference from secondary chemistry. (b) Early time OH sen-
sitivity; 15.5 ppm TBHP, 20 ppm methyl iodide, balance
Ar; initial reflected shock conditions: 1263 K, 1.28 atm;
S = (dXOH/dki)(ki/XOH), where ki is the rate coefficient
for reaction i.
of reaction (1) using methyl iodide as a precursor is
not as good as when using azomethane. The k-values
obtained using methyl iodide are systematically higher
than when using azomethane (see inset in Fig. 4). How-
ever, the difference is within the uncertainty and scatter
of the measurements. The uncertainty in the CH3I ex-
periments is estimated at ∼ 35%, slightly higher than
the uncertainty in the azomethane measurements. This
increase is primarily due to increased interference (see
Fig. 2b) from the following secondary reactions:
3CH2 + OH → CH2O + H
CH3I + OH → CH2I + H
3CH2 + CH3 → C2H4 + H
(11)
(12)
(13)
International Journal of Chemical Kinetics DOI 10.1002/kin

CH3 + OH AND CH3OH + Ar REACTION RATE COEFFICIENTS
Table III CH3OH + Ar → CH3 + OH + Ar: Rate
493
35
30
25
20
15
10
5
Experiment
Model best fit, k2a
2k2a, k2a/2
Coefficient Data
T5 (K)
P5 (atm)
k
(cm3 mol−1 s−1
)
2a
50.24 ppm CH3OH, balance Ar
2057
2100
1904
2151
2151
2033
2017
2298
2081
1.29
1.33
1.41
1.23
1.23
1.28
1.33
1.21
1.31
2.56 × 109
2.67 × 109
6.40 × 108
4.61 × 109
4.94 × 109
2.22 × 109
1.78 × 109
9.48 × 109
2.84 × 109
0
(a)
−40 −20
0
20 40 60 80 100 120
Time ( s)
CH3OH+AR
CH3+OH+AR
1CH2+H2O
The rate expressions for reactions (14a) and (14b) are
from GRI-Mech 3.0 [18], and yield a branching ratio,
k14a/(k14a + k14b), of ∼0.70 in the 1900–2300 K tem-
perature range. This is consistent with Baulch et al.
[9] who conclude that at high temperatures the pre-
dominant channel is reaction (14a). Owing to its small
sensitivity, the branching ratio chosen for reaction (14)
does not affect the current measurements of k2a.
A systematic error analysis taking into account
experimental and mechanism-induced contributions
yields an uncertainty of ∼ 30% in k2a at 2151 K
and 1.23 atm. The current measurements of k2a are
presented in Table III.
1.2
CH3+OH
2OH
CH3OH+AR
OH+CH3OH
O+H2O
1CH2+H2O+AR
CH3O+H2O
1.0
0.8
0.6
0.4
0.2
0.0
−0.2
−0.4
(b)
−10
0
10
20
30
40
50
Time ( s)
Figure 3 Example methanol decomposition experiment:
(a) OH concentration time history; solid black line: best-fit
RESULTS AND DISCUSSION
k
2a, dashed lines: factor of two variation in k2a. (b) OH
sensitivity; 50.24 ppm methanol, balance Ar; initial
reflected shock conditions: 2151 K, 1.23 atm; S
=
Figure 4 presents a comparison of the current measure-
ments of k1 with past work and theory. Evident in Fig. 4
is the relatively low scatter in the present experimen-
tal data. The current data are in reasonable agreement
(to within ∼50%) with the Bott and Cohen [5] mea-
surement at 1200 K and recent theoretical calculations
by Jasper et al. [3]. In the overlapping temperature
range, our measurements agree with the Krasnoperov
and Michael [6] data, but are about a factor of 2 higher
than the data of Srinivasan et al. [7]. The present work is
consistent with theory [3] that predicts a slight decrease
in k1 with an increase in the temperature in our exper-
imental regime. Within the scatter and uncertainty of
the current measurements, no significant pressure de-
pendence could be discerned in the overall rate—this
observation is consistent with Srinivasan et al. [7]. The
inset plot in Fig. 4 compares rate measurements using
two different methyl radical precursors, azomethane
and methyl iodide. The methyl iodide data are some-
what higher than the azomethane measurements, but as
pointed out earlier in the paper, the difference is within
the uncertainty of the current experiments. By virtue
(dXOH/dki)(ki/XOH), where ki is the rate coefficient for
reaction i.
Srinivasan et al. [7] was used for reaction (2b). It is per-
tinent to note that the k2b value reported in Srinivasan
et al. is for krypton as the third-body collider, whereas
the current measurements were carried out in argon.
However, the effect, on our k2a measurements, of the
difference in the collision efficiencies of argon and
krypton is negligible. This is because the current mea-
surements are only weakly sensitive to reaction (2b)
(see Fig. 3b), and hence large uncertainties can be tol-
erated in its rate coefficient (the latter also applies to
other secondary chemistry in the reaction system, OH
+ CH3OH, OH + OH, etc.). In the kinetic mechanism,
two product paths were considered for the reaction be-
tween OH and CH3OH [18],
OH + CH3OH → CH3O + H2O
(14a)
OH + CH3OH → CH2OH + H2O (14b)
International Journal of Chemical Kinetics DOI 10.1002/kin

494
VASUDEVAN ET AL.
1014
CONCLUSIONS
1015
The reaction between methyl and hydroxyl radicals
was studied in reflected shock wave experiments by
monitoring OH using narrow linewidth ring-dye laser
absorption at 306.7 nm. OH radicals were generated
by the rapid thermal decomposition of TBHP, while
azomethane and methyl iodide were used as CH3 radi-
cal precursors. Rate coefficients for the reaction of OH
with CH3 were inferred at elevated temperatures under
conditions of chemical isolation. Within estimated ex-
perimental uncertainty limits, the data are in agreement
with Bott and Cohen [5] and a recent theoretical study
by Jasper et al. [3]. The decomposition of methanol to
methyl and OH radicals was also studied. The current
measurements are in good agreement with Srinivasan
et al. [7] and with the master equation simulations of
Jasper et al. [3].
1013
1014
0.7
0.8
0.9
1013
1012
1000 K
2000 K
0.4
0.6
0.8
1.0
1.2
1000/T (K−1)
Figure 4 Arrhenius plot for CH3 + OH → Products:
solid squares, this work, azomethane as the CH3 precursor,
∼1.6 atm; open square, this work, azomethane as the CH3
precursor, ∼5 atm; open triangles, Krasnoperov and Michael
[6]; open circles, Srinivasan et al. [7] ∼700 Torr; solid cir-
cles, Srinivasan et al. [7] ∼250 Torr; open star, Bott and
Cohen [5]; solid black line, Jasper et al. [3] theory, 1.65 atm;
Inset, this work data with different methyl precursors: crossed
circles, methyl iodide as the CH3 precursor, ∼1.3 atm; solid
squares, azomethane as the CH3 precursor, ∼1.6 atm.
We thank Dr. David Davidson for his help and support. We
also thank Dr. Robin Walsh (University of Reading) and
Dr. Greg Smith (SRI International) for useful discussions on
the CH3 + OH reaction system.
1010
109
108
BIBLIOGRAPHY
1. De Avillez Pereira, R.; Baulch, D. L.; Pilling, M. J.;
Robertson, S. H.; Zeng, G. J Phys Chem A 1997, 101,
9681–9693.
2. Herbon, J. T.; Hanson, R. K.; Golden, D. M.; Bowman,
C. T. Proc Comb Inst 2005, 30, 955–963.
3. Jasper, A. W.; Klippenstein, S. J.; Harding, L. B.; Ruscic,
B. J Phys Chem A 2007, 111, 3932–3950.
4. Xia, W. S.; Zhu, R. S.; Lin, M. C.; Mebel, A. M. Faraday
Discuss 2001, 119, 191–205.
5. Bott, J. F.; Cohen, N. Int J Chem Kinet 1991, 23, 1017–
1033.
6. Krasnoperov, L. N.; Michael, J. V. J Phys Chem A 2004,
108, 8317–8323.
107 2500 K
1600 K
0.40 0.45 0.50 0.55 0.60 0.65
1000/T (K−1)
Figure 5 Arrhenius plot for CH3OH + Ar → CH3
+
OH + Ar: open squares, this work, ∼1.3 atm; solid circles,
Srinivasan et al. [7], ∼0.3–1.1 atm; solid line, Srinivasan
7. Srinivasan, N. K.; Su, M.-C.; Michael, J. V. J Phys Chem
A 2007, 111, 3951–3958.
8. Zervas, E.; Poulopoulos, S.; Philippopoulos, C. Fuel
2006, 85, 333–339.
et al. fit: k = 5.62 × 1015 exp(–30857/T ); solid gray line,
2a
Jasper et al. [3] theory, 1.3 atm.
9. Baulch, D. L.; Bowman, C. T.; Cobos, C. J.; Cox, R. A.;
Just, Th.; Kerr, J. A.; Pilling, M. J.; Stocker, D.; Troe,
J.; Tsang, W.; Walker, R. W.; Warnatz, J. J Phys Chem
Ref Data 2005, 34, 757–1399.
10. NIST Chemistry WebBook; available online at
http://webbook.nist.gov/chemistry. Accessed November
2007.
11. Vasudevan, V.; Davidson, D. F.; Hanson, R. K. Int J
Chem Kinet 2005, 37, 98–109.
12. Vasudevan, V.; Davidson, D. F.; Hanson, R. K. J Phys
Chem A 2005, 109, 3352–3359.
of their lower uncertainty, the azomethane experiments
(solid squares) are considered more reliable.
Figure 5 compares our k2a data with the recent mea-
surements of Srinivasan et al. [7] and with the theoret-
ical results of Jasper et al. [3]. The present measure-
ments have lower scatter than previous work. Within
the scatter of the Srinivasan et al. data and the uncer-
tainty in the current work, good agreement is observed.
Good agreement also is observed with the theoretical
results of Jasper et al.
International Journal of Chemical Kinetics DOI 10.1002/kin

CH3 + OH AND CH3OH + Ar REACTION RATE COEFFICIENTS
495
13. Jahn, F. P. J Am Chem Soc 1937, 59, 1761–1762.
14. Diels, O.; Koll, W. Justus Liebigs Ann Chem 1925, 443,
242–262.
15. Remmler, M.; Ondruschka, B; Zimmermann, G. J Prakt
Chem 1985, 327, 868–870.
16. Benson, S. W.; O’Neal, H. E. Kinetic Data on Gas
Phase Unimolecular Reactions, NSRDS-NBS 1970,
21.
17. Choo, K. Y.; Benson, S. W. Int J Chem Kinet 1981, 13,
833–844.
V. V.; Qin, Z. GRI-Mech 3.0. http://www.me.berkeley.
edu/gri mech/. Accessed November 2007.
19. Du, H.; Hessler, J. P.; Ogren, P. J. J Phys Chem 1996,
100, 974–983.
20. Herbon, J. T.; Hanson, R. K.; Golden, D. M.; Bowman,
C. T. Proc Comb Inst 2002, 29, 1201–1208.
21. Ruscic, B.; Feller, D; Dixon, D. A.; Peterson, K. A.;
Harding, L. B.; Asher, R. L.; Wagner, A. F. J Phys Chem
A 2001, 105, 1–4.
22. 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. J Phys Chem A 2002, 106, 2727–
2747.
18. Smith, G. P.; Golden, D. M.; Frenklach, M.; Moriarty,
N. W.; Eiteneer, B.; Goldenberg, M.; Bowman, C. T.;
Hanson, R. K.; Song, S.; Gardiner, W. C., Jr.; Lissianski,
International Journal of Chemical Kinetics DOI 10.1002/kin