Kinetics of OH radical reactions with methane in the temperature range 295-660 K and with dimethyl ether and methyl-tert-butyl ether in the temperature range 295-618 K

Article abstract of DOI:10.1021/jp012425t

The pulse laser photolysis/laser induced fluorescence (PLP-LIF) technique has been used to measure the rate constants of the reactions of OH radicals with dimethyl ether and methyl-tert-butyl ether. OH radicals were produced by photolysis of H₂

Full text of DOI:10.1021/jp012425t

4384
J. Phys. Chem. A 2002, 106, 4384-4389
Kinetics of OH Radical Reactions with Methane in the Temperature Range 295-660 K and
with Dimethyl Ether and Methyl-tert-butyl Ether in the Temperature Range 295-618 K
Am e´ lie Bonard, V e´ ronique Da e1 le, Jean-Louis Delfau, and Christian Vovelle*
Laboratoire de Combustion et Syst e` mes R e´ actifs, CNRS and UniVersit e´ d’Orl e´ ans, 1C, aVenue de la recherche
scientifique, 45071 Orleans Cedex 2, France
ReceiVed: June 26, 2001; In Final Form: February 14, 2002
The pulse laser photolysis/laser induced fluorescence (PLP-LIF) technique has been used to measure the
rate constants of the reactions of OH radicals with dimethyl ether and methyl-tert-butyl ether. OH radicals
were produced by photolysis of H
furnaces in order to obtain information on the temperature dependency of the rate constants in the domain of
95-660 K. A preliminary study of the reaction of OH with methane was carried out to control the experimental
setup. The Arrhenius expression obtained in the temperature range of 295-668 K, kOH+CH4 ) (5.65 ( 0.49)
2 2
O at λ ) 266 nm. The photolysis cell was heated by a small electric
2
-
21 3.01
×
10
T
exp[-(959 ( 36)/T], is in very good agreement with previous determinations (bimolecular rate
3
-1 -1
constant units: cm molecule s ; error limits ( 2σ). For the reactions of OH with ethers, the Arrhenius
-
20 2.85
expressions derived from our own results are kOH+DME(295-618 K) ) (3.02 ( 0.10) × 10
13)/T] and kOH+MTBE(297-616 K) ) (6.59 ( 0.43) × 10 exp([(499 ( 22)/T]. Combining these
data with those from previous experimental studies allows us to derive Arrhenius expressions in larger
T
exp[(618
-
19 2.40
(
T
-
19 2.46
temperature domains: kOH+DME(230-650 K) ) (4.59 ( 0.21) × 10
T
exp[(476 ( 14)/T] and
-
20 2.93
k
OH+MTBE(230-750 K) ) (1.58 ( 0.09) × 10 exp([(716 ( 18)/T].
T
Introduction
Concerning the thermal stability of the organic reagent, it is
likely, from experimental and modeling studies on the oxidation
There is a growing use of oxygenated compounds, especially
organic ethers, as fuel additives with beneficial effects such as
an increase of the octane number and a reduction of carbon
monoxide and particles emissions. However, these additives can
2
3
of DME and MTBE, that thermal degradation is negligible at
a temperature lower than 750 K. However, we considered that
it would be useful to perform preliminary measurements on the
reaction of OH radicals with methane, a reagent with a better
thermal stability than ethers. Also, this reaction has been
also enhance the exhaust emission of volatile organic compounds
1
(VOCs) such as formaldehyde. A better understanding of the
4
extensively studied in a large temperature domain with a very
influence of oxygenated fuel additives requires accurate data
on their consumption kinetics in automotive engines. Normal
combustion in an engine occurs essentially at high temperature
good agreement between several independent determinations.
Experimental Section
(between 1500 and 2000 K), and kinetic data should be
Figure 1 displays the main features of the experimental setup.
The reaction cell is made of quartz and is heated by independent
electrical furnaces in order to maintain a flat temperature profile
in the reaction zone. Temperature measurements along the
reaction cell axis showed that in the range 318-618 K the
temperature gradient is less than 1.0 K cm . These measure-
ments were performed with a type K thermocouple introduced
from the bottom of the reaction cell.
Reactants flow through a three concentric tubes assembly.
The central one is closed at its end and contains a chromel-
alumel thermocouple for temperature control. The molecular
reactant, diluted by helium, flows through the intermediate tube,
and an H2O2/He mixture flows through the external tube. Gas
flow rates are controlled by mass flow meters with an individual
accuracy of 3%.
determined in this range. However, data obtained in a lower
temperature range (from ambient to 800 K) are also of interest
because important processes such as knock take place in this
temperature domain.
In this work, the rate constants for the reactions of OH radicals
with dimethyl ether (DME) and methyl-tert-butyl ether (MTBE)
have been measured in the temperature range 295-618 K.
Measurements have been performed using the pulsed laser
photolysis/laser induced fluorescence (PLP-LIF) technique,
with a reaction cell heated electrically. OH radicals were
produced by photolysis of H2O2. The production of OH by
thermal decomposition of H2O2 could increase the background
signal and reduce the accuracy of the OH fluorescence measure-
ments. Specific measurements of the fluorescence signal in the
absence of the photolysis light showed that OH production by
thermal decomposition of H2O2 was completely negligible up
to 570 K. A progressive increase in the signal was observed at
higher temperatures, but the photolysis remains the dominant
process for the formation of OH up to 670 K.
-
1
The fourth harmonic of a Nd:YAG (GCR-100, Quanta-Ray)
laser is used for the production of OH radicals in the ground
2
state (X π (ν′′ ) 0)) by photolysis of H2O2 at λ ) 266 nm. The
OH concentration is monitored by laser-induced fluorescence
using a Nd:YAG pumped-frequency dye laser. OH excitation
2
2 +
at λ ) 282 nm corresponds to the X π (ν′′ ) 0) f A Σ
(ν′ ) 1) transition. The dye laser operates with Rodamine 590
*
To whom correspondence should be addressed. Phone: 33 2 38 25 54
8
2. Fax: 33 2 38 69 60 04. E-mail: vovel@cnrs-orleans.fr.
1
0.1021/jp012425t CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/04/2002

Kinetics of OH Radical Reactions
J. Phys. Chem. A, Vol. 106, No. 17, 2002 4385
Figure 1. Experimental setup.
Figure 2. Reaction OH + CH
4
. Evolution with time of the fluorescence
-
3
signal. (a) T ) 295 K; CH
4
concentration (molecule cm ), (b) 0.0;
in methanol. Both the photolysis and fluorescence lasers have
repetition rates of 10 Hz. Typical pulse energies, measured at
the exits of the reaction cell, are 10-45 mJ for the photolysis
laser and slightly less than 1 mJ for the fluorescence laser. A
time delay generator (Stanford Research System) synchronizes
the laser shots at the preset reaction time.
1
6
16
(
(
2) 4.05 × 10 ; (9) 6.95 × 10 . (b) T ) 619 K; CH
4
concentration
-3
15
16
molecule cm ), (b) 0.0; (2) 2.67 × 10 ; (9) 1.26 × 10 .
(50/50%) solution was concentrated by bubbling of helium
during 2 days.
A photomultiplier (PM) collects the radiation emitted in a
direction perpendicular to the plane defined by the two laser
beams. A filter with 310 nm peak transmission and 10 nm
Results and Discussion
2
+
bandwidth selects the signals corresponding to the A Σ
(a) OH + CH Reaction. The reaction of OH with methane
4
2
(
ν′ ) 0) f X π (ν′′ ) 0) transition at λ ) 308 nm and to
has been studied in the temperature range of 295-668 K at a
2
+
2
4
the A Σ (ν′ ) 1) f X π (ν′′ ) 1) transition at λ ) 314 nm.
The output signal of the PM is integrated over a preset time
period (gate width, 100 ns) by a Boxcar (EG&G, Princeton
Applied Research) and transmitted to a microcomputer. The
signal integration was repeated 128 times to increase the signal-
to-noise ratio.
total pressure of 1.33 × 10 Pa (100 Torr). Figure 2 shows that
the LIF signal (I ) decays exponentially in the whole domain
OH
of methane concentrations. The pseudo-first-order rate constants
(k′) were obtained by fitting the variations of ln(IOH) with time
to straight lines by using a weighted linear least-squares
7
procedure. As recommended by Cvetanovic et al., data were
2
The reactant concentration is derived from its partial pressure
in the reaction cell under the perfect gas assumption. To control
that rate constant measurements are conducted under pseudo-
first-order conditions, the OH radical concentration has been
estimated from the H2O2 concentration with a quantum ef-
weighted by wi ) 1/σi with σi ) σIOH/IOH. σIOH is the mean
standard deviation of the 128 OH fluorescence signal measure-
ments. The value of k′ obtained without methane was denoted
k0′ and corresponds to the contribution of two terms: the
consumption of OH by reaction with H2O2 (kH O [H2O2]) and
2
2
5
ficiency of 2 and an absorption cross section depending on
the diffusion of OH out of the detection zone (kd). External
consistency values were computed for the standard deviations
of k′ and k0′.
6
the temperature. The H2O2 concentration has been measured
at room temperature by absorption of the 213.9 nm line from a
zinc lamp. For these measurements, an absorption cell (1 m
length) was connected to the exit of the reaction cell. A band-
pass filter (214 nm peak transmission and 12 nm bandwidth)
isolated the 213.9 nm line. The transmitted light was detected
by a photodiode. All experiments on OH + DME and OH +
MTBE reactions have been done with [organic reactant]/[OH]
Figure 3 shows the variation of the net pseudo-first-order rate
(k′ - k′) with the methane concentration. A linear relationship
0
is observed for all temperatures studied, and rate constants of
the bimolecular reaction of OH with CH4 are derived from the
slope of each straight line obtained from data fitting. Data were
2
2
2 1/2
weighted by wi ) 1/σi with σi ) [σk′ + σk′ ] . Table 1 lists
o
4
ratios above 100. This ratio is considerably higher (>10 ) for
the values and their error limits taken as twice the standard
deviation. The range of OH and CH4 concentrations are also
mentioned in this table.
the reaction CH4 + OH to make the secondary reaction CH3 +
OH negligible.
The certified purity of Helium (Alphagaz) was 99.9995%.
Those of the reactants were 99.90% (methane, Alphagaz), 99%
The Arrhenius plot of the rate constants measured in this work
(Figure 4) exhibits a curvature, and a two step procedure was
used to determine the three parameters of a modified Arrhenius
expression: k ) AT exp(-E/T). In the first step, variations of
ln(k) with T were fitted to the expression y ) ln(k) ) ln(A1) +
(DME, Fluka), and 99.8% (MTBE, Aldrich). Both ethers, liquid
n
at ambient temperature, were purified by distillation prior to
mixing with helium in a storage bottle. The initial H2O2/H2O

4386 J. Phys. Chem. A, Vol. 106, No. 17, 2002
Bonard et al.
used with data weighted by wi ) 1/σzi2 and σz ) σk /ki. The
i
i
best fit so obtained corresponds to the expression
kOH+CH4 (this work) ) (5.65 ( 0.49) ×
-
21 3.01
exp[-(959 ( 36)/T] cm molecule s-1
3
-1
1
0
T
Kinetic data8
-12
previously measured with different tech-
niques (shock tube, flow reactor, and photolysis cell) have been
1
1
also plotted in Figure 4. Dunlop and Tully determined the
rate constant between 293 and 800 K. They mentioned an
outstanding agreement between their values and those measured
9
between 223 and 420 K by Vaghjiani and Ravishankara. From
both sets of data, they derive a modified Arrhenius expression:
Figure 3. Reaction OH + CH
tion of the net pseudo first-order rate constant. Temperature (K): (b)
95; (2) 367; (9) 464; ([) 619.
4
. Evolution with the methane concentra-
2
kOH+CH4 (Dunlop and Tully) )
-
20 2.58
exp(-1082/T) cm molecule s-1
3
-1
9
.65 × 10
T
4
Atkinson observed that this expression is also in excellent
agreement with the measurements performed by Bott and
8
10
Cohen at 1234 K, Finlayson-Pitts et al. (278-378 K), and
1
2
Mellouki et al. (233-343 K) and proposed to extend the
validity of the Dunlop and Tully expression to the temperature
range 233-1234 K.
Our measurements are also in excellent agreement with all
values considered by Atkinson, especially between 295 and 420
K, so that it can be assumed that error sources were minimized.
The main error sources, secondary reaction of OH with CH3
and reactions of OH with methane impurities such as alkenes,
would both lead to an overestimation of the rate constant of
the OH + CH4 reaction. Because of the fast increase of the
rate constant of the OH + CH4 reaction with the temperature,
this overestimation would be maximum at low temperature. In
the temperature range 600-668 K, the three parameter expres-
sion obtained in this work predicts values 15% larger than those
calculated by the Atkinson’s expression. This slight difference
can result from problems related to the thermal stability of H2O2.
However, it may be also considered that only one data set11
covers this temperature range in Atkinson’s recommendation.
When all data plotted in Figure 4, including very recent
Figure 4. Reaction OH + CH
4
. Arrhenius plot of the rate constant
and comparison with previous experimental determinations. (b, this
1
2
11
8
work; ×, Mellouki et al.; 3, Dunlop and Tully; 0, Bott and Cohen
9
10
O, Vaghjiani and Ravishankara; 4, Finlayson-Pitts et al.; [, Gierczak
13
13
et al.; modified Arrhenius equations: s, this work; - - -, Atkinson;
measurements performed by Gierczak et al. with the PLP-
-
- -, all data.)
LIF technique between 195 and 296 K, are taken into account,
the following three parameter expression is obtained in the
temperature range 195-1234 K:
TABLE 1: Ranges of Reactant Concentrations and Reaction
Rate Constants of the Reaction OH + CH
4
at a Total
4
Pressure of 1.33 × 10 Pa (100 Torr)
kOH+CH4 (all data of Figure 4) ) (4.09 ( 0.18) ×
temp
K)
range of OH
concentration
range of CH
concentration
4
b
k ( 2σ
a
(cm molecule s-1
3
-1
(
)
-21 3.04
exp[-(920 ( 13)/T] cm molecule s-1
3
-1
1
0
T
-
-
-
-
-
-
-
-
-
15
15
14
14
14
13
13
13
13
295
319
367
419
464
519
569
619
668
1.1-1.6
1.2-1.2
9.8-14.6
2.1-2.7
1.0-1.1
2.5-4.7
2.7-3.6
6.4-7.0
4.6-6.7
0.6-8.7
0.5-8.2
0.5-5.5
0.4-4.0
0.5-2.8
0.2-2.6
0.1-1.5
0.1-1.5
0.1-1.4
(6.23 ( 0.20) × 10
(9.16 ( 0.28) × 10
(2.19 ( 0.07) × 10
(4.26 ( 0.18) × 10
(7.83 ( 0.21) × 10
(1.33 ( 0.04) × 10
(2.05 ( 0.09) × 10
(3.06 ( 0.10) × 10
(4.22 ( 0.14) × 10
The largest difference with respect to Atkinson’s recommenda-
tion is only 5%.
(
b) OH + CH3OCH3 Reaction. This reaction has been
studied in the temperature range of 295-618 K at a total
pressure of 1.33 10 Pa (100 Torr). As for the reaction with
4
methane, exponential decays of the OH signal are observed over
at least three lifetimes. Figure 5 shows the variation of the net
pseudo-first-order reaction rate with the DME concentration,
for three temperatures. Straight lines are obtained, and a
weighted linear least-squares procedure is used to derive the
rate constants of the OH + DME reaction. Values obtained
and the ranges of OH and DME concentrations are listed in
Table 2.
a
011 molecule cm-3. b 1016 molecule cm-3
1
.
n1 ln(T) -E1/T by a nonlinear least-squares procedure. Data were
2
weighted by wi ) 1/σyi with σy ) σk /ki, and values of A1, n1,
i
i
and E1 were obtained from minimization of the sum S )
N
i)1
2
∑
wi(yi - ln(A1) - n1ln(Ti) + E1/Ti) . In the second step, A2
A partial decomposition of DME by the λ ) 266 nm radiation
of the photolysis laser is not expected to affect the accuracy of
the results because DME does not absorb radiations with wave-
and E2 values and their errors σA and σE were computed from
2
2
fitting of the data to the expression z ) ln(k/Tn
E2/T. In this second step, a linear least-squares procedure was
1
) ) ln(A2) -
1
4
lengths above 200 nm. However, the absence of photolysis

Kinetics of OH Radical Reactions
J. Phys. Chem. A, Vol. 106, No. 17, 2002 4387
Figure 5. Reaction OH + DME. Evolution with the DME concentra-
tion of the net pseudo-first-order rate constant. Temperature (K): (b)
2
97; (2) 433; (9) 618.
TABLE 2: Ranges of Reactant Concentrations and Reaction
Rate Constants of the Reaction OH + DME at a Total
4
Pressure of 1.33 × 10 Pa (100 Torr)
temp
K)
range of OH
concentration
range of DME
k ( 2σ
a
b
(cm molecule s-1
)
3
-1
(
concentration
-12
12
-12
2
2
2
3
3
4
4
5
6
6
95c
97
98
0.5-0.6
0.5
0.8-0.9
0.7-0.8
0.9-1.0
1.0
3.2-18.7
2.5-18.2
2.4-20.2
2.7-18.1
2.4-15.5
1.3-12.0
1.6-10.9
1.2-7.7
(2.70 ( 0.08) × 10
(2.69 ( 0.07) × 10
(2.67 ( 0.07) × 10
(2.85 ( 0.06) × 10
(3.34 ( 0.11) × 10
(4.07 ( 0.11) × 10
(5.19 ( 0.14) × 10
(6.21 ( 0.12) × 10
(7.54 ( 0.20) × 10
(7.31 ( 0.24) × 10
Figure 6. Reaction OH + DME. Arrhenius plot of the rate constant
and comparison with previous experimental determinations. (b, this
-
d
19
15
work; ×, Wallington et al.; 3, Wallington et al.; O, Tully and
-
-
-
-
-
12
12
12
12
12
17
69
33
99
67
18
17
20
21
Droege; 4, Perry et al.; +, Mellouki et al.; ], Arif et al.;
, Nelson et al.;16 modified Arrhenius equations: s, this work; - - -,
1
all data.)
1.3
2.4-2.6
2.7-2.8
1.3-1.4
The Arrhenius plot in Figure 6 includes all previous deter-
e
-12
12
17
1.1-7.1
-
minations and those from this work. Our values confirm the
curvature of the Arrhenius plot but less pronounced than that
from Arif et al.’s study. The two step procedure used for the
OH + CH4 reaction was applied to compute the parameters of
a modified Arrhenius expression in the temperature range 295-
18
0.5-4.5
a
011 molecule cm-3. b 1014 molecule cm-3. Pulse energy of the
photolysis laser multiplied by a factor of 2. Reactants flow velocity
changed by a factor of 0.5. Reactants flow velocity changed by a factor
c
1
d
e
of 2.
6
18 K:
of DME was verified by repeating one measurement at ambient
temperature with the pulse energy of the photolysis laser
kOH+DME (this work) ) (3.02 ( 0.10) ×
-
12
increased by a factor of 2. The value so obtained, 2.70 × 10
-20 2.85
exp[(618 ( 13)/T] cm molecule s-1
3
-1
1
0
T
3
-1 -1
cm molecule s , is very close (within the incertitude range)
to those obtained with the current laser pulse energy (2.69 ×
Using the same fitting procedure, a second modified Arrhe-
nius expression was derived from all values plotted in Figure
-
12
3
-1 -1
1
0
cm molecule s ). A second control was made to test
for a possible influence of the reactants jets velocity in the
concentric tubes. Results obtained at ambient temperature and
at 617 K show that the variations with respect to the values
measured at the nominal velocities are within the incertitude
ranges.
6
. In the temperature range 230-650 K, this expression is
kOH+DME (all data of Figure 6) ) (4.59 ( 0.21) ×
-19 2.46
exp[(476 ( 14)/T] cm molecule s-1
3
-1
1
0
T
Values obtained in this work have been compared to previous
experimental determinations of the rate constant of the OH +
However, marked variations in σ values are observed from
one group to another. These variations result essentially from
differences in the definition of σ which include sometimes
experimental uncertainties in addition to statistical uncertainties.
15,16
DME reaction performed using either relative
or absolute
methods.1
6-21
Most determinations cover a limited temperature
2
1
domain (240-440 K), and only one (Arif et al. ) gives
information on the temperature dependence of the rate constant
in a larger temperature domain (295-650 K). Arif et al.
observed that the extension of the temperature range at values
higher than 440 K clearly shows a marked curvature in the
Arrhenius plot. A nonlinear least-squares analysis leads to a
modified Arrhenius expression:
2
With data weighting by wi ) 1/σi , data with large σ values
exert only a small influence on the parameters of the Arrhenius
expression. Data fitting was also performed without weighting
and the following expression was obtained:
kOH+DME (all data of Figure 6) ) (1.17 ( 0.11) ×
-
18 2.33
exp[(423 ( 35)/T] cm molecule s-1
3
-1
1
0
T
kOH+DME (Arif et al.) ) (3.39 ( 13.8) ×
(
c) OH + (CH3)3COCH3 Reaction. If only one reaction
0-
24
T
(4.11(0.6)
exp[(1221 ( 252)/T] cm molecule s-1
3
-1
1
path has to be considered for the reaction with DME, OH +
CH3OCH3 ) H2O + CH3OCH2, the reaction with MTBE
involves two paths: (A) OH + (CH3)3COCH3 ) H2O + (CH3)3-
COCH2 and (B) OH + (CH3)3COCH3 ) H2O + CH2(CH3)2-
COCH3. Because the reaction products were not analyzed in
this work, the measurements of the rate constant refer to the
sum of the two reaction paths. These measurements have been
Because n ) 4.11 seems unreasonably high, Arif et al.
proposed a second expression with n fixed at 2.0:
kOH+DME (Arif et al.) ) (1.05 ( 0.10) ×
-
17 2.0
exp[(328 ( 32)/T] cm molecule s-1
3
-1
10
T

4388 J. Phys. Chem. A, Vol. 106, No. 17, 2002
Bonard et al.
Results obtained by using relative methods at room temper-
ature, by different groups, have been compared in Table 4. These
values are in reasonable agreement. Complementary kinetic
studies have been carried out to derive information on the
temperature dependency of the rate constant. Between 240 and
25
4
40 K, Wallington et al. obtained a linear Arrhenius plot with
a small positive temperature dependency of the rate constant.
From a linear least-squares analysis, these authors derived a
two parameters Arrhenius expression:
kOH+MTBE (Wallington et al.) ) (5.1 ( 1.6) ×
-
12
exp[-(155 ( 100)/T] cm molecule s-1
3
-1
1
0
Figure 7. Reaction OH + MTBE. Evolution with the MTBE
concentration of the net pseudo-first-order rate constant. Temperature
In the temperature range 246-314 K, Bennett and Kerr26 also
observed a small positive temperature dependency and derived
from their own experimental results the following Arrhenius
expression:
(K): (b) 316; (2) 498; (9) 616.
TABLE 3: Ranges of Reactant Concentrations and Reaction
Rate Constants of the Reaction OH + MTBE at a Total
4
Pressure of 1.33 × 10 Pa (100 Torr)
temp
K)
range of OH
concentration
range of MTBE
concentration
k ( 2σ
kOH+MTBE (Bennett et al.) ) (4.0 ( 1.3) ×
3
-1
(
(cm molecule s-1
)
-
12
exp[-(102 ( 70)/T] cm molecule s-1
3
-1
1
0
-
12
2
2
2
2
3
3
4
4
5
6
6
6
97
97
97
98
16
67
33
98
66
16
16
16
0.5-0.6
0.7
1.1-1.2
0.2
0.7-0.8
0.9-1.0
1.2-1.3
1.7-2.0
2.1-2.4
3.6-3.7
2.4-2.6
0.8-0.9
3.1-19.6
2.5-19.8
2.5-20.8
1.2-9.66
2.9-18.4
2.5-16.0
2.1-13.5
1.6-10.4
1.4-9.1
(3.15 ( 0.04) × 10
(3.01 ( 0.09) × 10
(3.08 ( 0.09) × 10
(2.96 ( 0.04) × 10
(3.13 ( 0.05) × 10
(3.71 ( 0.07) × 10
(4.47 ( 0.10) × 10
(5.26 ( 0.18) × 10
(6.40 ( 0.14) × 10
(7.27 ( 0.27) × 10
(7.34 ( 0.23) × 10
(7.42 ( 0.22) × 10
c
d
e
-12
-12
-12
and
kOH+MTBE (Bennett et al., evaluation) ) (5.2 ( 1.7) ×
-1
-
-
-
-
-
-
12
12
12
12
12
12
-12
exp[-(155 ( 100)/T] cm molecule s-1
3
1
0
when data from others groups are also taken into account.
From measurements in the temperature range of 230-371
K, a slight curvature in the Arrhenius plot is mentioned by T e´ ton
1.3-8.4
f
-12
-12
0.7-4.7
0.6-4.3
e
27
et al., and a three parameter Arrhenius expression is proposed:
a
011 molecule cm-3. b 1014 molecule cm-3. Pulse energy of the
photolysis laser multiplied by a factor of 2. Reactants flow velocity
changed by a factor of 0.5. Total pressure 6.65 kPa. Reactants flow
velocity changed by a factor of 2.
c
1
d
kOH+MTBE (T e´ ton et al.) ) (7.92 ( 0.40) ×
e
f
-17
T exp[(447 ( 32)/T] cm molecule s-1
2
3
-1
1
0
However, for comparison purpose, these authors also propose
an Arrhenius expression
performed in the temperature range of 297-616 K, at a total
4
pressure of 1.33 × 10 Pa (100 Torr). Here again, the
exponential decay of the OH signal is observed over three
lifetimes. Figure 7 shows that the net pseudo-first-order rate
constant increases linearly with the MTBE concentration. For
clarity purpose only, three temperature data have been plotted.
At each temperature, the rate constant is calculated by a
weighted linear least-squares analysis. Table 3 lists the values
obtained in this way and the ranges of OH and MTBE
concentrations.
kOH+MTBE (T e´ ton et al.) ) (5.03 ( 0.27) ×
10 exp[-(133 ( 30)/T] cm molecule s-1
-
12
3
-1
in very close agreement with Wallington’s and Bennett’s
expressions.
2
1
Arif et al. extended markedly the temperature domain
toward higher values (293-750 K) and observed a marked
curvature in the Arrhenius plot. From data fitting by a nonlinear
least-squares procedure, these authors derive a three parameters
expression:
The potential influence of either the pulse energy of the
photolysis laser or of the reactants velocity has been also tested
and found negligible as for DME. The effect of the total pressure
has also been examined at ambient temperature and at 616 K.
A reduction by a factor of 2 does not produce a significant
change in the rate constant.
As for the reaction with DME, most determinations of the
rate constant of the OH + MTBE reaction have been performed
either at room temperature or in limited temperature domains.
kOH+MTBE (Arif et al.) ) (1.11 ( 3.18) ×
17
(2.04(0.4)
exp[(266 ( 177)/T] cm molecule s-1
3
-1
1
0-
T
All data considered in this discussion and those obtained in
this work have been plotted in Figure 8. From our own results,
TABLE 4: Comparison of Relative Determinations of the Rate Constant of the OH + MTBE Reaction at Ambient
Temperature
reference
organics
k
OH+Ref
k ( 2σ
(cm molecule s-1
3
-1
)
(cm molecule s-1
3
-1
)
ref
T (K)
OH precursor
HONO
HONO
-
-
-
12
12
12
-12
2
2
2
2
2
2
2
95 ( 2
95 ( 2
98
98
98 ( 4
98
C
2
H
6
4
8.0 × 10
5.9 × 10
2.5 × 10
(2.6 ( 0.5) × 10
22
22
15
23
24
26
26
-12
n-C
n-C
n-C
n-C
H
H
H
H
14
10
10
12
(2.4 ( 0.4) × 10
-12
-12
-12
-12
-12
CH
CH
3
ONO
ONO or HONO
4
4
5
(3.24 ( 0.08) × 10
(2.99 ( 0.12) × 10
(2.98 ( 0.06) × 10
(2.84 ( 0.08) × 10
(3.09 ( 0.08) × 10
-
-
-
-
12
3
2.54 × 10
3.96 × 10
1.08 × 10
1.29 × 10
12
11
11
2 2 3
H O or CH ONO
CH
CH
3
ONO + NO
ONO + NO
C
C
2
H
5
OC
OC
2
2
H
H
5
98
3
2
H
5
5

Kinetics of OH Radical Reactions
J. Phys. Chem. A, Vol. 106, No. 17, 2002 4389
where several data sets are available. At higher temperatures,
our measurements are slightly larger than those of Dunlop
and Tully but differ from the Atkinson recommendation by
less than 5%.
For the OH + DME reaction, this study confirms the
curvature of the Arrhenius plot and provides new values in
agreement with both the low temperature determinations of
2
0
21
Mellouki et al. and the measurements of Arif et al. between
2
95 and 650 K.
Measurements performed in this work for the reaction OH +
MTBE reaction are in excellent agreement with previous data
2
7
measured between 295 and 371 K by T e´ ton et al. At higher
temperatures, our values are slightly smaller than those previ-
2
1
ously measured by Arif et al.
Acknowledgment. Whalid Mellouki and Georges Le Bras
are warmly acknowledged for their continuous interest in the
progress of this work. One of us (A.B.) is also very grateful to
“
Region Centre” for financial support.
References and Notes
Figure 8. Reaction OH + MTBE. Arrhenius plot of the rate constant
and comparison with previous experimental determinations. (b, this
(1) Guibet, J. C. Carburants et Moteurs; Nouvelle Edition, Editions
Technip: Paris, 1997.
(2) Curran, H. J.; Pitz, W. J.; Westbrook, C. K.; Dagaut, P.; Boettner,
J. C.; Cathonnet, M. Int. J. Chem. Kinet. 1998, 30, 229.
2
2
26
work; 0, Cox and Goldstone; O, Bennett and Kerr; ×, Wallington
2
5
15
23
27
et al.; 3, Wallington et al.; 4, Smith et al.; +, T e´ ton et al.; ],
(3) Glaude, P. A.; Battin-Leclerc, F.; Judenherc, B.; Warth, V.; Fournet,
2
1
24
Arif et al.; [, Picquet et al.; modified Arrhenius equations: s, this
R.; C oˆ me, G. M.; Scacchi, G. Combust. Flame 2000, 121, 345.
work; - - -, all data.)
(4) Atkinson R. J. Phys. Chem. Ref. Data 1997, 26 (2), 215.
(
5) Vaghjiani, G. L.; Ravishankara, A. R. J. Chem. Phys. 1990, 92,
the two steps least-squares analysis leads to the following
expression, in the temperature domain 297-616 K:
(2), 996.
(6) De More, W. B.; Sander, S. P.; Golden, D.; Hampson, R. F.; Kurylo,
M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina, C. J. NASA
JPL Pub. 1997, 97, 4.
(7) Cvetanovic, R. J.; Singleton, D. L.; Paraskevopoulos, G. J. Phys.
Chem. 1979, 83 (1), 50.
kOH+MTBE (this work) ) (6.59 ( 0.43) ×
-
19 2.40
exp[(499 ( 22)/T] cm molecule s-1
3
-1
10
T
(8) Bott, J. F.; Cohen, N. Int. J. Chem. Kinet. 1989, 21, 485.
(
9) Vaghjiani, G. L.; Ravishankara, A. R. Nature 1991, 350 (6317),
The solid line in Figure 8 corresponds to this expression.
Considering all data reported on this figure, the two steps least-
squares analysis with data weighting by wi ) 1/σi leads to a
4
06.
(10) Finlayson-Pitts, B. J.; Ezell, M. J.; Jayaweera, T. M.; Berko, H.
N.; Lai, C. C. Geophys. Res. Lett. 1992, 19 (13), 1371.
11) Dunlop, J. R.; Tully, F. P. J. Phys. Chem. 1993, 97, 11148.
12) Mellouki, A.; T e´ ton, S.; Laverdet, G.; Quilgars, A.; Le Bras, G. J.
Chim. Phys. 1994, 91, 473.
(13) Gierczak, T.; Talukdar, R.; Herndon, S. C.; Vaghjiani, G. L.;
Ravishankara, A. R. J. Phys. Chem. A 1997, 101, 3125.
(14) Calvert, J.; Pitts, J. N., Jr. Photochemistry; Wiley: New York, 1966.
15) Wallington, T. J.; Andino, J. M.; Skewes, L. M.; Siegl, W. O.;
Japar, S. M. Int. J. Chem. Kinet. 1989, 21, 993.
16) Nelson, L.; Rattigan, O.; Neavyn, R.; Sidebottom, H.; Treacy, J.;
2
(
(
second expression, in the temperature range of 230-750 K:
kOH+MTBE (all data of Figure 8) ) (1.58 ( 0.09) ×
-
20 2.93
exp[(716 ( 36)/T] cm molecule s-1
3
-1
10
T
(
The same procedure applied without data weighting leads to
(
Nielsen, O. J. Int. J. Chem. Kinet. 1990, 22, 1111.
(17) Perry, R. A.; Atkinson, R.; Pitts, J. N., Jr. J. Chem. Phys. 1977,
67, 611.
kOH+MTBE (all data of Figure 8) ) (3.97 ( 0.25) ×
-
21 3.15
exp[(752 ( 54)/T] cm molecule s-1
3
-1
10
T
(
(
18) Tully, F. P.; Droege A. T. Int. J. Chem. Kinet. 1987, 19, 251.
19) Wallington, T. J.; Liu, R.; Dagaut, P.; Kurylo, M. J. Int. J. Chem.
Conclusions
Kinet. 1988, 20, 41.
20) Mellouki, A.; T e´ ton, S.; Le Bras, G. Int. J. Chem. Kinet. 1995, 27,
91.
(21) Arif, M.; Dellinger, B.; Taylor, P. H. J. Phys. Chem. A 1997, 101,
436.
22) Cox, R. A.; Goldstone, A. Proceedings of the Second European
(
A heated PLP-LIF cell has been used to measure absolute
rate constants for reactions of OH radicals with DME and MTBE
in the temperature ranges 295-618 and 297-616 K, respec-
tively. The maximum temperature can be limited by thermal
degradation of either the H2O2 used as OH radicals precursor
or the organic reagent. From specific measurements of the OH
fluorescence signal in the absence of the photolysis light, it was
shown that the first problem is negligible up to 570 K and not
crucial up to 670 K. This point was also controlled by measuring
the rate constant of the extensively studied reaction OH + CH4.
Values obtained are in good agreement with previous determi-
nations, especially in the temperature range of 295-420 K
7
2
(
Symposium on the Physico-Chemical BehaViour of Atmospheric Pollutants,
D. Reidel Publishing Co.: Dordrecht, The Netherlands, 1981; p 112.
(
23) Smith, D. F.; Kleindienst, T. E.; Hudgens, E. E.; McIver, C. D.;
Bufalini, J. J. Int. J. Chem. Kinet. 1991, 23, 907.
24) Picquet, B.; Heroux, S.; Chebbi, A.; Doussin, J. F.; Durand-Jolibois,
(
R.; Monod, A.; Loirat, H.; Carlier, P. Int. J. Chem. Kinet. 1998, 30, 839.
(25) Wallington, T. J.; Dagaut, P.; Liu, R.; Kurylo, M. J. EnViron. Sci.
Technol. 1988, 22, 842.
(
26) Bennett, P. J.; Kerr J. A. J. Atmos. Chem. 1990, 10, 27.
(27) T e´ ton, S.; Mellouki, A.; Le Bras, G. Int. J. Chem. Kinet. 1996, 28,
291.