Kinetics of the C3H7O2 + NO reaction: Temperature dependence of the overall rate constant and the i-C3H7ONO2 branching channel

Article abstract of DOI:10.1021/jp026134b

The temperature dependence of the overall rate constant for the C₃H₇O₂ + NO reaction and the rate constant for the minor branching-channel resulting in the production of i-C₃H₇ONO₂ have been measured using the turbulent flow technique with high-pressure chemical ionization mass spectrometry for the detection of reactants and products. The temperature dependence of the overall rate constant for the C₃H₇O₂ + NO reaction was investigated between 298 and 213 K at 100 Torr pressure, and the data were fit by the following Arrhenius expression (with 2 standard deviation error limits indicated): 43_-0.9+1.0 x 10^-12 exp[(268 ± 56)/T] cm³ molecule^-1 s^-1. This expression agrees well with previous isomer-specific measurements of the n-C₃H₇ONO₂ and i-C₃O₂ + NO rate constants made at lower pressures. The temperature dependence of the rate constant for the minor reaction channel i-C₃H₇O₂ + NO → i-C₃H₇ONO₂ was investigated between 298 and 213 K at 100 Torr pressure. The following Arrhenius expression was determined for the minor channel: 4.9_-2.9^+5.3 × 10^-16 exp[(1380 ± 230)/T] cm³ molecule^-3 s^-1. The Arrhenius expressions for the overall rate and the i-C₃H₇ONO₂ producing channel indicate a branching ratio of about 0.006 at 298 K and 0.020 at 213 K at 100 Torr pressure, which is in good agreement with the predictions of a recently revised empirical model for alkyl nitrate branching ratios.

Full text of DOI:10.1021/jp026134b

3040
J. Phys. Chem. A 2003, 107, 3040-3047
Kinetics of the C3H7O2 + NO Reaction: Temperature Dependence of the Overall Rate
Constant and the i-C3H7ONO2 Branching Channel
Jessica M. Chow, Angela M. Miller, and Matthew J. Elrod*
Department of Chemistry, Oberlin College, Oberlin, Ohio 44074
ReceiVed: May 17, 2002; In Final Form: NoVember 22, 2002
The temperature dependence of the overall rate constant for the C3H7O2 + NO reaction and the rate constant
for the minor branching channel resulting in the production of i-C3H7ONO2 have been measured using the
turbulent flow technique with high-pressure chemical ionization mass spectrometry for the detection of reactants
and products. The temperature dependence of the overall rate constant for the C3H7O2 + NO reaction was
investigated between 298 and 213 K at 100 Torr pressure, and the data were fit by the following Arrhenius
+1.0
expression (with 2 standard deviation error limits indicated): 4.3
× 10-12 exp[(268 ( 56)/T] cm3
-0.9
molecule-1 s-1. This expression agrees well with previous isomer-specific measurements of the n-C3H7ONO2
and i-C3H7O2 + NO rate constants made at lower pressures. The temperature dependence of the rate constant
for the minor reaction channel i-C3H7O2 + NO f i-C3H7ONO2 was investigated between 298 and 213 K at
+5.3
100 Torr pressure. The following Arrhenius expression was determined for the minor channel: 4.9
×
-2.9
10-16 exp[(1380 ( 230)/T] cm3 molecule-1 s-1. The Arrhenius expressions for the overall rate and the
i-C3H7ONO2 producing channel indicate a branching ratio of about 0.006 at 298 K and 0.020 at 213 K at 100
Torr pressure, which is in good agreement with the predictions of a recently revised empirical model for
alkyl nitrate branching ratios.
Introduction
reaction 3 from the results of environmental chamber photolysis
studies of C3-C8 n-alkanes.2 The branching ratio [k3b/(k3a
+
Alkyl peroxy radicals (RO2) are important intermediate
species formed in the oxidation of alkanes in the atmosphere1
k3b)] is found to rise from 0.039 for the propyl case to 0.23 for
the n-octyl case at 298 K and 740 torr, thus indicating the
importance of the association channel 3b, particularly for the
larger alkyl peroxy radicals. The model can also be used to
predict the branching ratios for primary and tertiary nitrate
formation through the use of a scaling factor based on the
available results for experimental primary and tertiary systems.3
However, the experimental data is less extensive for these
species, and the predictions are expected to be less accurate.
In previous work, we extended the experimental measure-
ments to the C1 and C2 alkanes by measuring the overall rate
constant (k3) for the CH3O2 4 and C2H5O2 5 + NO reactions, as
well as values for the nitrate branching ratios. Our experimental
results for the C2H5ONO2 branching ratio were in only fair
agreement with the predictions from the Atkinson model.
Therefore, the goal of this work is to extend our experimental
measurements to the C3 system to assess whether the C2H5ONO2
inconsistencies are due to a specific inaccuracy in the Atkinson
model3 for this primary nitrate, or whether there might exist a
more general systematic discrepancy between the experimental
methods employed (which would be revealed by a differing
result for the C3H7ONO2 branching ratio).
RH + OH f R + H2O
(1)
(2)
R + O2 + M f RO2 + M
Ozone levels in the atmosphere are directly affected by RO2
reactions, and the identity of the dominant RO2 reactions are
dependent on the levels of the nitrogen oxides (NOx). Under
high NOx conditions (generally, lower tropospheric urban
conditions), RO2 reactions lead to the production of ozone, the
most deleterious constituent of photochemical smog
RO2 + NO f RO + NO2
NO2 + hν (λ < 420 nm) f NO + O
O + O2 + M f O3 + M
(3a)
(4)
(5)
RO2 can also be temporarily removed from ozone production
cycles by the formation of reservoir species such as the alkyl
nitrates
Reaction 1 leads to two possible alkyl radical isomers for
the case of the C3H8 + OH reaction. It has been shown that the
distribution of alkyl radicals isomers at 295 K is about 75%
i-C3H7 and 25% n-C3H7.6 There have been several previous
kinetics investigations of the overall rate of reaction of C3H7O2
+ NO.7-11 It has been well established that the reaction
RO2 + NO + M f RONO2 + M
(3b)
Because reaction 3b removes the radical chain carrier RO2 from
the ozone production cycle, it has a direct effect on the ozone-
producing efficiency of the RH/NOx system.
Atkinson and co-workers have developed an empirical
relationship for the yield of the secondary organic nitrates from
C3H7O2 + NO f C3H7O + NO2
(6a)
* To whom correspondence should be addressed. Phone: (440) 775-
6583. Fax: (801) 697-4917. E-mail: matthew.elrod@oberlin.edu.
is the predominant channel. In particular, the more recent work
10.1021/jp026134b CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/03/2003

Kinetics of the C3H7O2 + NO Reaction
J. Phys. Chem. A, Vol. 107, No. 17, 2003 3041
Figure 1. Experimental apparatus.
of Howard and co-workers have established that the overall rate
constant is identical for the two possible propyl peroxy isomers,
n-C3H7O2 and i-C3H7O2, with a value of 9.0 × 10-12 cm3
molecule-1 s-1 at 298 K and pressures near 1 Torr.8,9
The minor nitrate-forming channel for the C3H7O2 + NO
reaction is
Torr and at a range of temperatures extending to those found
in the lower stratosphere using a turbulent flow (TF) tube
coupled to a high pressure chemical ionization mass spectrom-
eter (CIMS). It has been previously shown that the TF technique
can be used to accurately determine the rate constants of
reactions at pressures ranging from 50 to 760 Torr and at
temperatures as low as 180 K.14,15 As in our previous kinetics
studies of CH3O2 + NO4 and C2H5O2 + NO5 using the coupled
TF-CIMS approach, we are able to directly access atmospheric
pressure and temperature conditions and sensitively monitor
many of the relevant reactants and products for the C3H7O2 +
NO reaction. In the C2H5O2 + NO work, the TF-CIMS kinetics
technique was used in the determination of the temperature-
dependent overall rate constant and the first measurement of
the C2H5ONO2 producing channel for this reaction,5 thus
improving on a previous estimate for the upper limit of this
process at room temperature and atmospheric pressure.12 In this
article, we describe a similar temperature-dependent kinetics
investigation of the overall rate constant and the measurement
of the i-C3H7ONO2 producing channel (the dominant propyl
nitrate-forming reaction) rate constant for the i-C3H7O2 + NO
reaction.
C3H7O2 + NO + M f C3H7ONO2 + M
(6b)
In 1982, using an environmental smog chamber, Atkinson and
co-workers first measured the secondary propyl nitrate (i-
C3H7ONO2) branching ratio [k6b/(k6a + k6b)] to be 0.036 at 299
K and 735 torr.12 Based on more recent estimates for the propyl
radical distribution resulting from the reaction of propane and
OH, Arey et al. recently adjusted the secondary propyl nitrate
branching ratio to a value of 0.039.2 More importantly, new
measurements of the C6-C8 systems were performed, and new
parameters for the temperature and pressure dependent empirical
model were also established.2 The new model predicts the
secondary propyl nitrate branching ratio at 299 K and 735 torr
to be 0.041, in good agreement with the experimental results.
Although the overall rate constants for the different C3H7O2
isomers in reaction 6a were found by Howard and co-workers
to be identical, the work of Atkinson and co-workers has shown
that there is a clear isomer dependence in the nitrate branching
ratios for the entire C3-C8 series. Therefore, it is important to
use methods that are selective for the specific isomers in the
nitrate branching ratio studies.
Bertman et al. have proposed that the alkyl nitrate branching
ratios can be used in regional atmospheric models that determine
the air mass age from observations of the relative concentration
of alkyl nitrate to the parent alkane.13 Although their method
was found to be self-consistent for the secondary C4 and C5
nitrates, Bertman et al. found that their model systematically
underpredicted measured ethyl and propyl nitrate levels by a
substantial amount. There are two potential explanations for this
result: (1) the experimental alkyl nitrate branching ratios are
too low or (2) ethyl and propyl nitrates are also formed after
the decomposition of larger alkoxy radicals, leading to higher
amounts of these alkyl nitrates than predicted from the
atmospheric levels of ethane and propane. In previous work,
using our experimental results for the C2H5ONO2 branching ratio
and the Bertman et al. model, we supported the contention that
most C2H5ONO2 present in the atmosphere is not a result of
ethane oxidation chemistry.5
Experimental Section
Turbulent Fast Flow Kinetics. A schematic of the experi-
mental apparatus is presented in Figure 1 and is similar to that
used in our previous study of C2H5O2 + NOx.5 The flow tube
was constructed with 2.2 cm i.d. Pyrex tubing and was 100 cm
in total length. A large flow of nitrogen carrier gas (ap-
proximately 30 STP L min-1) was injected at the rear of the
flow tube. The gases necessary to generate C3H7O2 were
introduced though a 10 cm long 12.5 mm diameter sidearm
located at the rear of the tube. NO was added via an encased
movable injector. The encasement (made from corrugated Teflon
tubing) was used so that the injector could be moved to various
injector positions without breaking any vacuum seals, as well
as to prevent ambient gases from condensing on cold portions
of the injector. A fan-shaped Teflon device was placed at the
end of the injector in order to enhance turbulent mixing. The
polonium-210 alpha-emitting ionization source was placed
between the temperature regulated flow tube and the inlet to
the CIMS. Most of the flow tube gases were removed at the
CIMS inlet by a 31 L s-1 roughing pump. All gas flows were
monitored with calibrated mass flow meters. The flow tube
pressure was measured upstream of the ionization source using
a 0-1000 Torr capacitance manometer. The temperature was
In this article, we describe our investigation of the kinetics
of the C3H7O2 + NO reaction conducted at pressures near 100

3042 J. Phys. Chem. A, Vol. 107, No. 17, 2003
Chow et al.
determined at both the entrance and exit points of the temper-
ature regulated region of the flow tube using Cu-constantan
thermocouples.
Overall Rate Constant Determination. C3H7O2 (all mol-
ecules mentioned are mixed isomers, unless otherwise indicated)
was generated using the following reactions:
L min-1 flow of helium (99.999%), which had passed through
a silica gel trap immersed in liquid nitrogen, with a 0.4 STP
mL min-1 flow of a 1% H2 (99.9%)/He mixture. To generate
i-C3H7, the hydrogen atoms were then injected into a sidearm
and mixed with an excess of C3H6 (CP grade, ∼2 × 1014
molecule cm-3) in order to ensure that no hydrogen atoms were
introduced into the main flow. i-C3H7O2 was then produced by
the addition of a large excess of O2 to the main flow (99.995%;
>5.0 × 1017 molecule cm-3) at the intersection of the sidearm
and the main flow tube. Absolute i-C3H7O2 concentrations were
determined by the titration of NO to produce i-C3H7O and NO2
C3H8 + Cl f C3H7 + HCl
(7)
(8)
C3H7 + O2 + M f C3H7O2 + M
(k7 ) 1.37 × 10-10 cm3 molecule-1 s-1 and k8 ) 1.1 × 10-11
cm3 molecule-1 s-1 at 100 Torr; all rate constant values quoted
are for 298 K unless otherwise indicated).16,17 Chlorine atoms
were produced by passing a dilute Cl2/He mixture through a
microwave discharge produced by a Beenakker cavity operating
at 50 W. The dilute Cl2/He mixture was obtained by combining
a 2.0 STP L min-1 flow of helium (99.999%), which had passed
through a silica gel trap immersed in liquid nitrogen, with a
0.5-5.0 STP mL min-1 flow of a 1% Cl2 (99.9%)/He mixture.
To generate C3H7, the chlorine atoms were then injected into a
sidearm and mixed with an excess of C3H8 (CP grade, ∼2 ×
1014 molecule cm-3) in order to ensure that no chlorine atoms
were introduced into the main flow. C3H7O2 was then produced
by the addition of a large excess of O2 (99.995%; >2.0 × 1016
molecule cm-3) just downstream of the production of C3H7.
Absolute C3H7O2 concentrations (needed to ensure pseudo-first-
order kinetics conditions and for branching ratio modeling) were
determined by the titration of C3H7O2 with NO (see C3H7ONO2
Branching Channel Reactant Preparation section). To ensure
pseudo-first-order kinetics conditions, the main flow tube
C3H7O2 concentrations ranged from 5.0 to 10 × 1010 molecules
cm-3. The C3H8 + Cl reaction also leads to a similar i-C3H7-
dominated isomer distribution,18 as is the case for the C3H8 +
OH reaction. Thus, our overall rate constant measurements
represent a weighted average value for the two isomers.
For the overall rate determination for the C3H7O2 + NO
reaction, NO was added to the flow reactor as a 1% mixture in
N2 through the movable injector. To ensure the absence of NO2
impurities, NO was passed through a silica gel trap held at -78
C. Negligible NO2 concentrations were observed via CIMS
methods.
i-C3H7O2 + NO f i-C3H7O + NO2
(11a)
(k11a ) 9.0 × 10-12 cm3 molecule-1 s-1 9
) and subsequent
calibration of the NO2 mass spectrometer signal. 0.5 % NO2/
N2 mixtures were used as standard samples for mass spectro-
metric calibration. Computer modeling of the reactions occurring
during the titration indicated that not all of the i-C3H7O2 initially
present was converted to the stable species, NO2. This is mainly
due to reactions of i-C3H7O2 with other species in the chemical
system (which will be addressed in the discussion section).
Therefore, the i-C3H7O2 concentrations were calculated from
the NO2 concentrations by explicitly determining the conversion
factor (usually ∼0.9) from computer modeling using the specific
experimental conditions. As mentioned in the previous section,
NO was purified in situ by passing it through a silica gel trap
held at -78 C. For this study, i-C3H7O2 concentrations ranged
from 3.0 to 7.0 × 1011 molecule cm-3
.
i-C3H7ONO2 Branching Channel Measurements. In these
studies, the production of i-C3H7ONO2 from reaction 11b
i-C3H7O2 + NO f i-C3H7ONO2
(11b)
was directly monitored over a reaction time of ∼30 ms.
Computer modeling was used to extract the rate constant k11b
from the observed production of i-C3H7ONO2 and initial
concentrations of all relevant chemical species. As will be
discussed in the Results and Discussion section, both the time
profile of the i-C3H7ONO2 production and the absolute amount
of i-C3H7ONO2 produced over the longest reaction time were
used as constraints on the determination of k11b; therefore, it
was critical that the absolute i-C3H7ONO2 concentrations were
determined accurately. Standard samples of i-C3H7ONO2 (Al-
drich, 96%) for mass spectrometric calibration were com-
mercially obtained and prepared by drawing an appropriate
amount of i-C3H7ONO2 vapor from the liquid sample, followed
by mixing with N2 to make 0.5% i-C3H7ONO2 mixtures
appropriate for mass spectrometric calibration.
For the low-temperature studies, liquid-nitrogen-cooled sili-
cone oil was used as the coolant for the jacketed flow tube.
Nitrogen carrier gas was precooled by passing it through a
copper coil immersed in a liquid N2 reservoir followed by
resistive heating. The temperature was controlled in the reaction
region to within 1 K. The pressure was maintained at ap-
proximately 100 Torr in order to achieve optimum instrument
performance. Pressures below 100 Torr were found to lower
the ionization efficiency, and pressures above 100 Torr were
found to lower the radical production efficiency.
To aid in the detection of the i-C3H7ONO2 product, higher
i-C3H7O2 were used for the branching ratio measurements than
in the determination of the bimolecular rate constant (because
i-C3H7O2 is the limiting reagent). The major complication in
the branching ratio determination is from the reaction of the
i-C3H7O radical (formed in reaction 11a) with NO2
i-C3H7ONO2 Branching Channel Reactant Preparation.
i-C3H7O2 was selectively generated using the following reac-
tions:
i-C3H7O + NO2 + M f i-C3H7ONO2 + M
(12)
H + C3H6 f i-C3H7
(9)
(k12 ) 3.3 × 10-11 cm3 molecule-1 s-1 at 100 Torr).20 Reaction
12 is therefore a potential source of i-C3H7ONO2, and experi-
mental conditions must be designed to minimize this relatively
fast reaction. By keeping O2 concentrations very high, computer
modeling indicates that the reaction
i-C3H7 + O2 + M f i-C3H7O2 + M
(10)
(k9 ) 2.0 × 10-12 cm3 molecule-1 s-1 and k10 ) 1.1 × 10-11
cm3 molecule-1 s-1 at 100 Torr).17,19 H atoms were produced
by passing a dilute H2/He mixture through a microwave
discharge produced by a Beenakker cavity operating at 50 W.
The dilute H2/He mixture was obtained by combining 2.0 STP
i-C3H7O + O2 f CH3C(O)CH3 + HO2
(13)

Kinetics of the C3H7O2 + NO Reaction
J. Phys. Chem. A, Vol. 107, No. 17, 2003 3043
(k13 ) 6.5 × 10-15 cm3 molecule-1 s-1 20,21
)
will consume most
of the i-C3H7O produced by reaction 11a and minimize
production of i-C3H7ONO2 by reaction 12. This reaction is
slightly less efficient (5.2 × 10-15 cm3 molecule-1 s-1 at 213
K) at the lowest temperatures used in this study.
To drive the production of i-C3H7ONO2 from reaction 12
down to very low levels, O2 was added to the flow tube at
concentrations of ∼5.0 × 1017 molecule cm-3. Additional
experiments were performed with O2 concentrations of 1.0 ×
1018 molecule cm-3 in order to investigate potential interferences
from this reaction.
Chemical Ionization Mass Spectrometric Detection. Posi-
tive ion chemical ionization schemes (with H+(H2O)n as the
reagent ion) were used to detect i- and n-C3H7O2 and
i-C3H7ONO2, and negative ion chemical ionization schemes
-
(with SF6 as the reagent ion) were used to detect NO2 with
the quadrupole mass spectrometer. H+(H2O)n (with the n ) 4
species usually accounting for more than 80% of the total ion
signal) was produced in the ion source by passing a large O2
flow (8 STP L min-1) through the polonium-210 alpha-emitting
ionization source (with H2O impurities being sufficiently
-
abundant to produce adequate quantities of reagent ions). SF6
was produced in the ion source by passing a large N2 flow (9
STP L min-1) and 1.0 STP mL min-1 of 10% SF6/N2 through
the ionization source. The commercial ionization source con-
sisted of a hollow cylindrical (69 by 12.7 mm) aluminum body
with 10 mCurie (3.7 × 108 disintegrations s-1) of polonium-
210 coated on the interior walls.
Ions were detected with a quadrupole mass spectrometer
housed in a two-stage differentially pumped vacuum chamber.
Flow tube gases (neutrals and ions) were drawn into the front
chamber through a charged 0.1 mm aperture. The ions were
focused by three lenses constructed from 3.8 cm i.d. and 48
cm o.d. aluminum gaskets. The front chamber was pumped by
a 6 in. 2400 L s-1 diffusion pump. The gases entered the rear
chamber through a skimmer cone with a charged 1.0 mm orifice
which was placed approximately 5 cm from the front aperture.
The rear chamber was pumped by a 250 L s-1 turbomolecular
pump. Once the ions passed through the skimmer cone, they
were mass filtered and detected with a quadrupole mass
spectrometer.
Figure 2. Pseudo-first-order C3H7O2 decay curves for the C3H7O2 +
NO reaction at 100 Torr, 298 K, and 1170 cm s-1 velocity.
-
NO2 was detected using a previously reported SF6 chemical
ionization scheme
NO2 + SF6- f NO2- + SF6
(16)
(k16 ) 1.4 × 10-10 cm3 molecule-1 s-1).23 We found that
i-C3H7ONO2 reacts very quickly with H+(H2O)4
C3H7ONO2 + H+(H2O)4 f C3H7ONO2‚H+(H2O)3 + H2O
(17)
such that high sensitivity could be achieved for the detection
of this important product. The detection sensitivity of i-C3H7ONO2
was estimated to be about 100 ppt at 100 Torr.
Chemical Ionization Schemes. Because the chemical ioniza-
tion reactions are not expected to be sensitive to the isomeric
structure of the neutral species under study and the mass
spectrometric method is incapable of distinguishing between
isomers, the ion-molecule reactions are discussed below in an
isomer-neutral fashion. To perform pseudo-first-order kinetics
studies of the overall rate of reaction 6, a chemical ionization
scheme for C3H7O2 is required. In our previous study of the
CH3O2 + NO reaction,4 we developed the following chemical
ionization detection scheme for CH3O2 based on the computa-
tional thermodynamic prediction22 that CH3O2 has a higher
proton affinity than H2O
Results and Discussion
Overall Rate Constant Determination. Bimolecular rate
constants were obtained via the usual pseudo-first-order ap-
proximation method, using NO as the excess reagent. Typical
C3H7O2 decay curves as a function of injector distance are
shown in Figure 2 for the NO kinetics measurements. The first-
order rate constants obtained from fitting the C3H7O2 decay
curves were plotted against [NO] in order to determine the
bimolecular rate constant, as shown in Figure 3. This approach
for determining bimolecular rate constants assumes that devia-
tions from the plug flow approximation (molecular velocities
are equal to the bulk flow velocity) are negligible. Under the
conditions present in our turbulent flow tube (Reynold’s number
>2000), Seeley et al. estimated that these deviations result in
apparent rate constants which are at most 8% below the actual
values.14 A portion of the plug flow deviation can be traced to
the diffusivity contribution to the flow continuity equation.
Because both eddy diffusion and molecular diffusion contribute
roughly equally to the diffusivity term, it is possible that
turbulent flow kinetics studies using molecules with smaller
diffusion coefficients (such as C3H7O2 in this case, as compared
to the H atom studies carried out by Seeley et al.) might lead
CH3O2 + H+(H2O)4 f CH3OOH+(H2O)3 + H2O
(14)
Because the proton affinity of C3H7O2 is expected to be very
similar to CH3O2, we successfully attempted to detect C3H7O2
with the analogous reaction
C3H7O2 + H+(H2O)4 f C3H7OOH+(H2O)3 + H2O (15)
To perform branching ratio measurements for reaction 11b,
chemical ionization schemes are needed for NO2 (to determine
the i-C3H7O2 concentration) and for the product i-C3H7ONO2.

3044 J. Phys. Chem. A, Vol. 107, No. 17, 2003
Chow et al.
Figure 4. Arrhenius plot of the temperature dependence of the overall
rate constant for the C3H7O2 + NO reaction.
Figure 3. Determination of overall bimolecular rate constant for the
C3H7O2 + NO reaction from data in Figure 2.
TABLE 1: Overall Rate Constant Data for the C3H7O2 +
NO Reaction at 100 Torr Pressure
a
velocity
Reynolds
number
k6
T (K)
(cm s-1
)
(10-12 cm3 molecule-1 s-1
)
298
298
298
288
288
288
273
273
263
263
253
253
243
243
243
243
233
233
233
223
223
213
213
1160
1150
1170
1250
1280
1300
1060
1060
970
991
974
957
906
923
916
916
960
955
955
961
970
2190
2220
2200
2500
2580
2560
2290
2290
2200
2330
2500
2470
2480
2470
2450
2450
2580
2730
2730
2970
2930
3260
3260
11.2 ( 1.9
10.5 ( 0.7
9.8 ( 1.6
11.4 ( 0.9
11.9 ( 0.9
11.3 ( 1.6
11.9 ( 0.8
10.4 ( 0.7
11.7 ( 1.4
12.2 ( 0.9
11.9 ( 0.7
12.9 ( 1.0
12.1 ( 1.4
11.9 ( 1.8
13.3 ( 1.9
13.3 ( 0.2
15.1 ( 1.2
14.2 ( 0.7
13.9 ( 0.5
13.8 ( 1.3
15.4 ( 1.1
14.7 ( 3.4
15.3 ( 1.3
966
966
Figure 5. Observed production of i-C3H7ONO2 from reaction 6b
(squares) above the i-C3H7ONO2 background level (circles) as a function
of injector distance. A least squares fit to the background level was
performed, and the dotted lines represent the 2σ level. This data set
was obtained under the following conditions: P ) 100 Torr, T ) 223
K, velocity ) 1180 cm s-1, Reynold’s number ) 3660, [i-C3H7O2]0 )
a Stated error is 2σ.
to larger plug flow deviations. However, because the flow is
modified by the use of a device (the “turbulizer” described in
the Experimental Section) to enhance turbulent mixing, the plug
flow approximation might be more exact for our conditions than
a numerical simulation of the basic turbulent flow dynamics
might suggest. Nontheless, it continues to be important to
confirm the plug flow validity of the turbulent flow method
via inspection of the pseudo-first-order decays and by com-
parison to other experimental methods. The other likely
systematic errors in the determination of rate constants are likely
to occur in the measurements of gas flows, temperature, detector
signal, and pressure. Considering such sources of error, we
3.40 × 1011 molecule cm-3, [NO]0 ) 4.56 × 1012 molecule cm-3
.
estimate that rate constants can be determined with an accuracy
of (30% (2σ).
We performed several determinations of the rate constant at
100 Torr and 298 K for the mixed isomer C3H7O2 + NO
reaction (see Table 1 for a complete list of experimental
conditions and measured rate constants) and arrived at the mean
value of k ) (10.5 ( 1.4) × 10-12 cm3 molecule-1 s-1; the
uncertainty represents the two standard deviation statistical errors
in the data and is not an estimate of systematic errors. Our room-
temperature constant is in excellent agreement with previous

Kinetics of the C3H7O2 + NO Reaction
J. Phys. Chem. A, Vol. 107, No. 17, 2003 3045
TABLE 2: Kinetics Parameters for Branching Ratio Determinationp
A
k3o00
k3∞00
(cm3 s-1
Ea/R
(K)
(cm6 s-1
molecule-2
(cm3 s-1
molecule-1
reaction
molecule-1
)
)
m
)
n
a
i-C3H7O2 + NO f i-C3H7O + NO2
i-C3H7O2 + NO f i-C3H7ONO2
2.7 × 10-12
-360
fitted
parameters
2240
b
2 i-C3H7O2 f 2 i-C3H7O + O2
1.3 × 10-12
9.7 × 10-13
4.7 × 10-13
7.5 × 10-13
1.0 × 1015
1.4 × 10-14
8.9 × 10-12
2 i-C3H7O2 f CH3C(O)CH3 + i-C3H7OHb
2240
c
i-C3H7O2 + NO2 f i-C3H7O2NO2
-620
-700
8807
210
-397
d
i-C3H7O2 + HO2 f i-C3H7OOH + O2
e
i-C3H7O f CH3CHO + CH3
f
i-C3H7O + O2 f CH3C(O)CH3 + HO2
i-C3H7O + NO f i-C3H7ONOg
h
i-C3H7O + NO2 f i-C3H7ONO2
2.0 × 10-27
4
3
3.3 × 10-11
1.8 × 10-12
1
1.7
CH3 + O2 f CH3O2
4.50 × 10-31
i
j
CH3O2 + NO f CH3O + NO2
9.2 × 10-13
3.5 × 10-12
4.8 × 10-11
-600
-250
-250
i
HO2 + NO f OH + NO2
i
HO2 + OH f H2O + O2
OH + OH f H2O2
6.2 × 10-31
7.0 × 10-31
2.5 × 10-30
1
2.6
4.4
2.6 × 10-11
3.6 × 10-11
1.6 × 10-11
0
0.1
1.7
i
OH + NO f HONOi
OH + NO2 f HNO3
i
OH+ CH3C(O)CH3 f CH3C(O)CH2 + H2Ok
OH + i-C3H7OH f i-C3H6OH + H2Ol
OH + C3H6 f C3H6OHm
1.25 × 10-12
2.7 × 10-12
4.86 × 10-12
1.2 × 10-11
1.0 × 10-11
561
-190
-504
0
C3H6OH + O2 f CH3CH(O2)CH2OHn
CH3CH(O2)CH2OH + NO f CH3CH(O)CH2OH + NO2
o
0
a Eberhard, J.; Villalta, P. W.; Howard, C. J. J. Phys. Chem. 1996, 100, 993. b Wallington, T. J.; Dagaut, P.; Kurylo, M. J. Chem. ReV. 1992, 92,
667. c From C2H5O2 + NO2 value, Ranschaert, D. L.; Schneider, N. J.; Elrod, M. J. J. Phys. Chem. A. 2000, 104, 5758. d From C2H5O2 + HO2
value, DeMore, W. B.; Sander, S. P.; Howard, C. J.; Ravishankara, A. R.; Golden, D. M.; Kolb, C. E.; Hampson, R. F.; Kurylo, M. J.; Molina, M.
J. Chemical Kinetics and Photochemical Data for Use in Stratospheric Modeling, JPL Publication 97-4; Jet Propulsion Laboratory: Pasadena, CA,
1997. e First order rate constant, Heicklen, J. AdV. Photochem. 1988, 14, 177. f Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr, J.
A.; Rossi, M. J.; Troe, J. Summary of EValuated Kinetic Data for Atmospheric Chemistry (http://www.iupac-kinetic.ch.cam.ac.uk); IUPAC
Subcommittee on Gas Kinetic Data Evaluation for Atmospheric Chemistry: Cambridge, U.K., 2001. g Fittschen, C.; Frenzel, A.; Imrik, K.; Devolder,
P. Int. J. Chem. Kinet. 1999, 31, 860. h k∞300 from Mund, C.; Fockenberg, C.; Zellner, R. Ber. Bunsenges. Phys. Chem. 1998, 102, 709, temperature
dependence from C2H5O + NO2 , DeMore, W. B.; Sander, S. P.; Howard, C. J; Ravishankara, A. R.; Golden, D. M.; Kolb, C. E.; Hampson, R. F;
Kurylo, M. J.; Molina, M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric Modeling, JPL Publication 97-4; Jet Propulsion
Laboratory: Pasadena, CA, 1997. i DeMore, W. B.; Sander, S. P.; Howard, C. J; Ravishankara, A. R.; Golden, D. M.; Kolb, C. E.; Hampson, R.
F; Kurylo, M. J.; Molina, M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric Modeling, JPL Publication 97-4; Jet Propulsion
Laboratory: Pasadena, CA, 1997. j Scholtens, K. W.; Messer, B. M.; Cappa, C. D.; Elrod, M. J. J. Phys. Chem. A 1999, 103, 4378. k Le Calve, S.;
Hitier, D.; Le Bras, G.; Mellouki, A. J. Phys. Chem. A 1998, 102, 4579. l Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr, J. A.;
Rossi, M. J.; Troe, J. Summary of EValuated Kinetic Data for Atmospheric Chemistry (http://www.iupac-kinetic.ch.cam.ac.uk); IUPAC Subcommittee
on Gas Kinetic Data Evaluation for Atmospheric Chemistry: Cambridge, U.K., 2001. m Tsang, W. J. Phys. Chem. Ref. Data 1991, 20, 221. n Miyoshi,
A.; Matsui, H.; Washida, N. J. Phys. Chem. 1990, 94, 3016. o Average alkyl peroxy + NO rate constant, Wallington, T. J.; Dagaut, P.; Kurylo, M.
J.; Chem. ReV. 1992, 92, 667. p Pressure independent rate constants calculated from k(T) ) Ae-E /RT. Pressure dependent rate constants calculated
a
2
-1
from k([M],T) ) {k0(T)[M]/[1 + (k0(T)[M]/k∞(T))]} × 0.6{1+[log
where k0(T) ) k300(T/300)-n and k∞(T) ) k300(T/300)-m
.
(k (T)[M]/k (T))] }
10
0
∞
0
∞
studies.7-11 Because of the excellent linearity of the pseudo-
first-order decays and the agreement with previous rate constant
determinations for this method, we conclude that plug flow
approximation holds well for the C3H7O2 + NO system under
our flow conditions.
calculating the nitrate branching ratios, we did not pursue the
measurement of the isomer-specific overall rate constants.
i-C3H7ONO2 Branching Ratio Determination. We were
able to observe the production of very small concentrations of
i-C3H7ONO2 (∼4.0 × 109 molecule cm-3) over the reaction
time (∼30 ms), which we have positively identified as coming
from reaction 11b. In Figure 5, the i-C3H7ONO2 rise (squares)
has been overlaid on the background signal (circles) to
demonstrate that the i-C3H7ONO2 rise is significantly larger than
the (2σ level of the background signal. Under the experimental
conditions shown in Figure 5, kinetics modeling shows that the
i-C3H7O + NO2 side reaction can produce concentrations of
i-C3H7ONO2 that are on the order of the detection level of the
instrument (∼5.0 × 108 molecule cm-3) and are less than the
actual i-C3H7ONO2 observed (∼4.3 × 109 molecule cm-3).
Table 2 contains a list of the reactions used in the modeling.
An inspection of Table 2 indicates that only one side reaction
directly produces i-C3H7ONO2
Temperature Dependence of the Overall Rate Constant
(k6) for C3H7O2 + NO. We performed several measurements
at 100 Torr pressure and at temperatures between 298 and 213
K in order to establish the temperature dependence of the rate
constant for conditions relevant to the upper troposphere. The
rate constant increased by approximately 40% as the temperature
was lowered over this range. From the data listed in Table 1
and plotted in Figure 4, we obtained the Arrhenius expression
+1.0
k(T) ) 4.3
× 10-12 exp[(268 ( 56)/T] cm3 molecule-1 s-1
.
-0.9
Within the stated uncertainties of the experiments, our temper-
ature dependence is in agreement with the results of Howard
and co-workers for their measurement of the temperature
dependence of the overall rate for n-C3H7O2 + NO and i-C3H7O2
+ NO reactions, which were made about 1 Torr pressure.8;9
Because of this good agreement with specific isomer C3H7O2
+ NO experiments of Howard and co-workers and because our
main intent in measuring the temperature dependent overall rate
constant was to provide an internally consistent method for
i-C3H7O + NO2 + M f i-C3H7ONO2 + M
(12)
However, this reaction is significant in that the reactants are
the products of the dominant reaction channel for i-C3H7O2 +
NO. As mentioned in the Experimental Section, oxygen was

3046 J. Phys. Chem. A, Vol. 107, No. 17, 2003
Chow et al.
TABLE 3: Branching Ratio Data for the i-C3H7O2 + NO Reaction at 100 Torr Pressure
velocity
Reynold’s
no.
[i-C3H7O2]0
[NO]0
k11b
branching ratio
T (K)
(cm s-1
)
(1011 molecule cm-3
)
(1011 molecule cm-3
)
(10-14 cm3 molecule-1 s-1
)
k
11b/(k11a + k11b)
298
298
298
298
298
298
298
298
298
298
298
298
298
298
263
253
253
243
243
233
233
233
233
233
223
223
223
223
223
213
213
213
213
1280
1280
1280
1280
1280
1300
1300
1300
1300
1300
1300
1300
1300
1300
1160
1120
1120
1100
1100
1180
1110
1110
1270
1270
1180
1180
1290
1290
1290
1280
1280
1440
1440
2450
2450
2450
2450
2450
2470
2470
2470
2470
2470
2470
2470
2470
2470
2710
2800
2800
2900
2860
3310
3250
3250
3670
3670
3660
3630
3840
3840
3840
4120
4120
4580
4580
6.9
6.9
6.9
6.9
6.9
3.3
3.3
3.3
3.3
3.3
3.3
3.3
3.3
3.3
3.3
2.7
2.7
3.4
3.4
4.1
5.0
5.0
5.0
5.0
3.4
3.4
5.3
5.3
5.3
5.0
5.0
5.0
5.0
36
55
46
36
36
35
46
55
55
33
46
46
52
52
38
51
41
46
46
50
31
17
28
21
46
46
35
28
22
33
33
29
20
5.4
2.7
6.8
4.5
6.8
6.3
3.6
3.6
4.5
4.5
3.6
3.6
5.4
7.2
13
7.8
13
24
23
25
17
11
14
14
0.006
0.003
0.008
0.005
0.008
0.007
0.004
0.004
0.005
0.005
0.004
0.004
0.006
0.008
0.012
0.007
0.012
0.020
0.019
0.020
0.013
0.009
0.011
0.011
0.030
0.026
0.025
0.013
0.016
0.012
0.014
0.020
0.021
41
35
34
18
22
18
21
29
31
added at higher concentrations in the branching ratio experi-
ments so that the following reaction was favored
(the temperature dependence of the C2H5O + NO2 reaction is
used as an estimate in the modeling calculations). Propagat-
ing the 10% value through our model indicates that this
uncertainty has a minor effect (less than 5%) on our fitted value
for k11b. However, because the lower temperature measurements
might be subject to higher uncertainty, we estimate the total
error to be approximately 50% for the branching ratio measure-
ments.
i-C3H7O + O2 f CH3C(O)CH3 + HO2
(13)
which effectively scavenges i-C3H7O from the system. As
described in the Experimental Section, two sets of experiments
were done at O2 concentrations of 5.0 × 1017 and 1.0 × 1018
molecule cm-3 to examine potential interferences from this
reaction. Both sets yielded the same results, indicating that the
production of i-C3H7ONO2 from reaction 12 was minimized
under the high O2 concentrations used in the branching ratio
experiments.
We performed several measurements of the i-C3H7ONO2-
producing branching ratio at temperatures between 213 and 298
K in order to establish the temperature dependence of the rate
constant k11b for conditions relevant to the upper troposphere.
From the data listed in Table 3 and presented in Figure 6, the
Computer modeling was used to extract a rate constant for
reaction 11b by fitting the observed i-C3H7ONO2 production.
The actual time profile of the i-C3H7ONO2 production, as well
as the total production of i-C3H7ONO2 over the total reaction
time were used as constraints in the fitting process. The model
input included the concentrations of C3H7O2, NO, and all
precursors. Table 3 contains a list of the initial conditions and
calculated rate constants (k11b) for the branching ratio experi-
ments. As stated earlier, we believe that the relevant experi-
mental uncertainty for a rate constant determination with this
apparatus is on the order of 30%, although the uncertainty may
be somewhat higher because the branching ratio measurements
are being made very close to the detection level of the
instrument. In addition, the branching ratio method requires the
added step of titration and calibration experiments needed to
determine [C3H7O2]0. The uncertainty in the rate constant for
reaction 12 is expected to play the largest role in the propagation
of rate constant uncertainties through our fitting process. The
room-temperature uncertainty in k12 is only 10%,21 but the
temperature dependence of k12 has not been directly measured
following Arrhenius expression is obtained for the experi-
+5.3
ments performed at 100 Torr total pressure: 4.9
× 10-16
-2.9
exp[(1380 ( 230)/T] cm3 molecule-1 s-1. Again, the uncertainty
represents the two standard deviation statistical error in the data
and is not an estimate of systematic errors (which are discussed
above). The branching ratios [k11b/(k11a + k11b)] reported in Table
3 were calculated using the Arrhenius parameters for the overall
rate constant (k11a + k11b) for i-C3H7O2 + NO given in Table
2. The branching ratios range from about 0.006 at 298 K to
0.02 at 213 K, indicating the increased importance of the
i-C3H7ONO2-producing channel at lower temperatures. Because
the i-C3H7ONO2-producing channel is almost certainly pressure-
dependent, it would be of interest to attempt to establish the
pressure dependence of the rate constant. However, because the
measurements made at 100 Torr were barely within the detection
limit of the instrument, it was not surprising that we were unable
to obtain sufficient sensitivity to pursue experiments at pressures
other than 100 Torr. It would also be of interest to pursue the
kinetics of the n-C3H7ONO2-producing reaction. However, we

Kinetics of the C3H7O2 + NO Reaction
J. Phys. Chem. A, Vol. 107, No. 17, 2003 3047
measurements of the temperature dependence of the overall rate
constant for the reaction of C3H7O2 with NO. Our measurement
of the temperature dependence of the overall rate constant at
100 Torr for the C3H7O2 + NO reaction was found to be in
good agreement with previous measurements performed at lower
pressures.8,9 The branching ratio [k11b/(k11a + k11b)] at 100 Torr
pressure for i-C3H7ONO2 formation was determined to be about
0.006 at 298 K and was found to increase to about 0.02 at 213
K. The temperature-dependent rate constant k11b was accurately
predicted by extrapolating from a model based on organic nitrate
formation from C3-C8 hydrocarbon systems.2 The present work
is useful for (a) obtaining the i-C3H7ONO2 yield data for
conditions representative of the upper troposphere (210 K and
100 torr pressure) and (b) when the present 298 K and 100 torr
pressure yield is combined with previous atmospheric pressure
data at 299 K, constraints concerning the pressure dependence
are placed on theoretical models of alkyl nitrate formation from
the RO2 + NO reactions.
Acknowledgment. The authors acknowledge experimental
assistance from Nick Schneider and Andy Huisman. This
material is based upon work supported by the National Science
Foundation under Grant No. 0196205.
Figure 6. Arrhenius plots of the temperature dependence of the rate
constant of the i-C3H7ONO2 branching channel for the present work
(data points and solid line) and from the predictions of the Atkinson
model (dotted line) at 100 Torr pressure.
References and Notes
(1) Brasseur, G. P.; Orlando, J. J.; Tyndall, G. S. Atmospheric
Chemistry and Global Change; Oxford University Press: New York, 1999.
(2) Arey, J.; Aschmann, S. M.; Kwok, E. S. C.; Atkinson, R. J. Phys.
Chem 2001, 105, 1020-1027.
(3) Carter, W. P. L.; Atkinson, R. J. Atmos. Chem. 1989, 8, 165.
(4) Scholtens, K. W.; Messer, B. M.; Cappa, C. D.; Elrod, M. J. J.
Phys. Chem. A 1999, 103, 4378.
are not aware of a discharge flow method to allow for the
selective generation of n-C3H7O2.
In Figure 6, the temperature-dependent i-C3H7ONO2-produc-
ing rate constant (k11b) obtained in the present work is compared
to the predictions from the Atkinson model2 at a total pressure
of 100 Torr. The model predicts a 298 K, 100 Torr branching
ratio of about 0.009, which is within the estimated uncertainty
of our experimental value of 0.006. The pressure dependence
of the i-C3H7ONO2 branching channel observed from the present
100 Torr results and the 735 Torr results of Atkinson and co-
workers is consistent with this reaction being in the third-order
limit in this pressure range. All in all, the performance of
empirical model is an impressive extrapolation, given that the
pressure and temperature dependence has been derived from
experimental results for the larger C5-C7 systems. However, it
is again worth noting that our previous results for the C2H5ONO2
branching ratios were in somewhat poorer agreement with the
Atkinson model results.5
In our previous work on the C2H5ONO2 branching ratio for
the C2H5O2 + NO reaction, we used our branching ratio results
and the Bertman et al. model13 described in the Introduction to
provide further support for the conclusion that field measure-
ments of C2H5ONO2 were not consistent with ethane oxidation
chemistry as the sole source of this species.5 In fact, the
calculation indicated that ethane was probably responsible for
less than 10% of the C2H5ONO2 present in the atmosphere. The
measurements presented here serve to provide further confidence
in the Atkinson model used by Bertman et al. for the calculation
of i-C3H7ONO2 from the i-C3H7O2 + NO reaction and, thus,
support the conclusion that some atmospheric i-C3H7ONO2 is
produced via the oxidation of larger hydrocarbons.13;24
(5) Ranschaert, D. L.; Schneider, N. J.; Elrod, M. J. J. Phys. Chem. A
2000, 104, 5758.
(6) Droge, A. T.; Tully, F. P. J. Phys. Chem. 1986, 90, 1949.
(7) Eberhard, J.; Howard, C. J. J. Phys. Chem. 1997, 101, 3360-3366.
(8) Eberhard, J.; Howard, C. J. Int. J. Chem. Kinet. 1996, 28, 731-
740.
(9) Eberhard, J.; Villalta, P. W.; Howard, C. J. J. Phys. Chem. 1996,
100, 993-997.
(10) Peeters, J.; Vertommen, J.; Langhans, I. Ber. Bunsen-Ges. Phys.
Chem. 1992, 96, 431-436.
(11) Adachi, H.; Basco, N. Int. J. Chem. Kinet. 1982, 14, 1243-1251.
(12) Atkinson, R.; Aschmann, S. M.; Carter, W. P. L.; Winer, A. M.;
Pitts, J. N., Jr. J. Phys. Chem. 1982, 86, 4563-4569.
(13) Bertman, S. B.; Roberts, J. M.; Parrish, D. D.; Buhr, M. P.; Goldan,
P. D.; Kuster, W. C.; Fehsenfeld, F. C.; Montzka, S. A.; Westberg, H. J.
Geophys. Res. 1995, 100, 22805.
(14) Seeley, J. V.; Jayne, J. T.; Molina, M. J. Int. J. Chem. Kinet. 1993,
25, 571.
(15) Seeley, J. V. Experimental Studies of Gas-Phase Radical Reactions
Using the Turbulent Flow Tube Technique; Massachusetts Institute of
Technology: Cambridge, MA, 1994.
(16) DeMore, W. B.; Sander, S. P.; Howard, C. J.; Ravishankara, A.
R.; Golden, D. M.; Kolb, C. E.; Hampson, R. F.; Kurylo, M. J.; Molina,
M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling, JPL Publication 97-4; Jet Propulsion Laboratory: Pasadena, CA,
1997.
(17) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr, J.
A.; Rossi, M. J.; Troe, J. IUPAC Subcommittee on Gas Kinetic Data
EValuation for Atmospheric Chemistry; 1999.
(18) Tschuikow-Roux, E.; Yano, T.; Niedzielski, J. J. Chem. Phys. 1985,
82, 65.
(19) Harris, G. W.; Pitts, J. N. J. Chem. Phys. 1982, 77, 3994.
(20) Fittschen, C.; Frenzel, A.; Imrik, K.; Devolder, P. Int. J. Chem.
Kinet. 1999, 31, 860.
(21) Mund, C.; Fockenberg, C.; Zellner, R. Ber. Bunsen-Ges. Phys.
Chem. 1998, 102, 709.
(22) Messer, B. M.; Stielstra, D. E.; Cappa, C. D.; Scholtens, K. W.;
Elrod, M. J. Int. J. Mass Spectrom. 2000, 197, 219.
(23) Huey, L. G.; Hanson, D. R.; Howard, C. J. J. Phys. Chem. 1995,
99, 5001.
(24) Flocke, F.; Volz-Thomas, A.; Buers, H. J.; Patz, W.; Garthe, H. J.;
Kley, D. J. Geophys. Res. Atmospheres 1998, 103, 5729.
Conclusions
The results presented here represent the first temperature
dependence measurements of the i-C3H7ONO2-producing rate
constant from the i-C3H7O2 + NO reaction, as well as