Kinetics of the Reaction between Ethylperoxy Radicals and Nitric Oxide

Article abstract of DOI:10.1021/jp9607935

The kinetic of the C2H5O2 + NO -> C2O5O + NO2 reaction is examined by two real-time techniques.NO consumption and NO2 formation are measured using transient diode laser absorption, whereas ethylperoxy loss and ethylnitrite formation are monitored by time-resolved UV spectrometry.Simultaneous fits of the NO and NO2 concentration versus time profiles yield rate constants consistent with the results of the C2H5O2 and C2H5ONO measurements.The combined data yield a rate constant of k5 = (3.1_-1.0^+1.5)E-12 e ^(330+/-11)/T cm³ s^-1 over the 220-355 K temperature range.The small negative temperature dependence is consistent with the accepted mechanism of the reaction proceeding through a C2H5O2NO adduct.

Full text of DOI:10.1021/jp9607935

1
2374
J. Phys. Chem. 1996, 100, 12374-12379
Kinetics of the Reaction between Ethylperoxy Radicals and Nitric Oxide
M. Matti Maricq* and Joseph J. Szente
Research Laboratory, Ford Motor Company, P.O. Box 2053, Drop 3083, Dearborn, Michigan 48121
X
ReceiVed: March 14, 1996; In Final Form: May 7, 1996
The kinetics of the C
consumption and NO
loss and ethylnitrite formation are monitored by time-resolved UV spectroscopy. Simultaneous fits of the
NO and NO concentration versus time profiles yield rate constants consistent with the results of the C
and C ONO measurements. The combined data yield a rate constant of k
2
H
5
O
2
+ NO f C
2
H
5
O + NO
2
reaction is examined by two real-time techniques. NO
2
formation are measured using transient diode laser absorption, whereas ethylperoxy
2
2 5 2
H O
+
1.5
1.0
-12 (330(110)/T
2
-1
H
5
5
) (3.1- ) × 10
e
3
cm s over the 220-355 K temperature range. The small negative temperature dependence is consistent
with the accepted mechanism of the reaction proceeding through a C H O NO adduct.
2 5 2
I. Introduction
timescale of the ensuing C2H5O2 + NO reaction by adding
sufficient quantities of ethane and oxygen, respectively; thus,
the generation of ethylperoxy radicals is assumed instantaneous.
The reaction between organic peroxy radicals and nitric oxide
is of central importance to the oxidation of hydrocarbon
emissions and to the generation of photochemical smog.
Atmospheric degradation generally commences, for simple
organic molecules, by OH attack and then O2 addition to
produce a peroxy radical. Its fate depends on the local
composition of the atmosphere, in particular to the NO, NO2,
and HO2 concentrations and to their respective rate constants
for reaction with RO2. The reactions with NO2 and HO2
generally produce peroxynitrates and hydroperoxides, respec-
tively. These compounds act as temporary reservoirs with the
radicals regenerated photolytically, by thermal decomposition,
or by OH reaction. Peroxyacetylnitrate is an important example
that, by sequestering NO2, provides a mechanism for its regional
transport. Further degradation of the hydrocarbon continues
with the RO2 + NO reaction. This is a key step in photochemi-
cal smog formation, since photolysis of the NO2 that is formed
regenerates nitric oxide and liberates an oxygen atom, which
reacts with O2 to form ozone.
2
The photolysis pulse, of approximately 300 mJ and 1 cm cross-
sectional area, travels longitudinally through the reaction vessel,
which is 51 cm in length and 3.2 cm in diameter, creating radical
1
4
-3
concentrations in the range of (5-10) × 10 cm . A typical
gas mixture consists of 0.4 Torr Cl2, 3 Torr C2H6, 0.1-0.4 Torr
NO, and 20-100 Torr O2, with nitrogen added to bring the
total pressure to about 100 Torr.
The reaction is examined in real time by two previously
6
described techniques: transient IR absorption and time-resolved
7
UV spectroscopy. In both cases the probe beam is antiparallel
with the photolysis beam. IR radiation provided by Pb-salt
diode lasers is used to monitor the loss of NO and the production
of NO2 during the course of the reaction. Vibration-rotation
lines in the P branch of the V ) 0 f 1 transition near 1850
-1
cm are used to probe NO, whereas absorption lines in the R
-1
branch of the ν3 asymmetric stretch at about 1630 cm are
used to detect NO2. After passing through the reaction cell,
the IR light is directed through a monochromator to remove
scattered light and impinges a HgCdTe detector having a
response time of about 200 ns. The IR laser frequency is
stabilized by passing a portion of the output through a reference
Despite the key importance of the title reaction to understand-
ing the atmospheric fate of ethane and of other compounds that
produce ethylperoxy radicals as degradation intermediates, the
rate constant for this reaction is not well established. Early
measurements by Adachi and Basco and by Plumb et al. differ
1
2
3
gas cell and locking the output to the desired NO (or NO2)
-
12
3
-1
absorption line. Laser jitter is reduced by using a microphone
by a factor of 3. The higher value of 8.9 × 10
cm s
to pick up the mechanical vibrations of the coldhead, which
are due to the compressor, and allowing the photolysis laser to
fire only during quiet periods between these vibrations.
recorded by Plumb et al. appears to be supported by the recent
4
5
measurements by Sehested et al. and Da e¨ le et al. Although
the rate constant has been assumed not to vary much with
temperature, the dependence remains unknown since the only
available data are at 295 K.
The detector is an ac coupled device whose output is
proportional only to the change in light intensity, I(t); thus, the
absolute intensity of the probe beam must be added to the signal
in order to convert the transient change in absorbance into a
concentration change. The latter quantity is obtained using
Beer’s law; thus,
II. Experimental Section
Ethylperoxy radicals are generated in a flowing gas stream
by the 351 nm excimer laser photolysis of molecular chlorine
in a Cl2/C2H6/NO/O2/N2 mixture and the following sequence
of reactions:
1
Ib + I(t)
[
X] )
ln
(3)
t
(
)
σ l
I0
X
Cl + C H f C H + HCl
(1)
(2)
2
6
2
5
where X represents NO or NO2, σX is its optical cross section,
l is the path length, I0 is the intensity of the IR beam with no
NO (or NO2) present, and Ib is the IR intensity prior to the laser
photolysis (i.e., NO is present in the cell). The required IR
cross sections, which are sensitive to temperature and pressure,
are determined separately for each experiment by measuring
C H + O + M f C H O + M
2
5
2
2
5
2
Reactions (1) and (2) are made rapid (t1/2 = 200 ns) on the
*
Author to whom correspondence is addressed.
Abstract published in AdVance ACS Abstracts, June 15, 1996.
X
S0022-3654(96)00793-9 CCC: $12.00 © 1996 American Chemical Society

Reaction between Ethylperoxy Radicals and Nitric Oxide
J. Phys. Chem., Vol. 100, No. 30, 1996 12375
the absorbance of a known concentration of NO (or NO2) under
the experimental conditions.
Monitoring the changes in ethylperoxy and ethylnitrite
concentrations is accomplished by substituting a D2 lamp for
the diode laser. This lamp provides broad-band UV radiation
in the spectral range of 190-350 nm. After passing through
the reaction cell, the light is dispersed by a 0.32 m monochro-
mator with a resolution of approximately 3 nm and recorded
by a gated diode array detector. The gate width is 4 µs. For
each photolysis laser pulse the lamp intensity is recorded
immediately prior to the pulse and at a delay time ranging from
3
to ∼100 µs after the pulse. These provide, respectively, the
I0 and I(t) intensities from which the time-resolved absorbance
spectra are calculated.
The absorbances are converted to concentrations by fitting
them to reference spectra of the principal UV-absorbing species,
namely, C2H5O2, (C2H5ONO-2NO + NO2), C2H5O2NO2, and
HNO via
Abs(λ,t) ) lσ (λ) C (t)
(4)
∑i
i
i
where l is the path length, σi(λ) is the wavelength-dependent
cross section of species i, and Ci(t) is its time-dependent
concentration. Because NO and NO2 absorb only weakly over
the 190-300 nm range, the spectra of these molecules are
combined with the ethylnitrite spectrum in order to reduce the
number of fitting parameters in eq 4. The C2H5O2, C2H5ONO,
and NO2 reference spectra are obtained from ref 7. Spectra of
HNO, C2H5O2NO2, and NO were measured as part of the
present investigation. HNO was produced by the photolysis of
Cl2 in the presence of CH2O and NO. C2H5O2NO2 was formed
by Cl2 photolysis in the presence of ethane, oxygen, and NO2.
The absolute intensities were calibrated against the spectrum
of C2H5O2 prepared under identical experimental conditions.
Gas flows are regulated by Tylan flow controllers, except
for the Cl2 flow which is set by a needle valve. Their
concentrations are determined by measuring the rate of pressure
increase for each component gas as it flows into a fixed volume.
These vary by only a few percent over the course of the
experiment. The cell is kept at a preset temperature by
recirculating ethanol or ethylene glycol/water through a jacket
surrounding the cell. The gases are preconditioned to the
temperature setpoint prior to entering the cell. Their temperature
is measured by thermocouples placed at the entrance and exit
of the cell. Cl2 (9.7% in He) and NO (certified 10.1% in N2)
are obtained from Matheson. The supplier’s analysis of the NO
mixture was verified by difference IR spectroscopy against pure
NO, which yielded a concentration of 10.2 ( 0.2%. N2
Figure 1. Transient IR absorption measurements of NO loss and NO2
formation for three initial NO concentrations. The smooth lines represent
best fits of the reaction model in Table 1 to the data, with k
5
treated as
an adjustable parameter.
Ethylperoxy radicals also react with each other,
C H O + C H O f 2C H O + O
2
(6a)
(6b)
2
5
2
2
5
2
2
5
C H O + C H O f C H OH + CH CHO + O
2
2
5
2
2
5
2
2
5
3
However, owing to the excess of NO employed in these
1
,9
experiments and to the small self-reaction rate constant, this
pathway for C2H5O2 removal is of negligible consequence.
Reaction (5) suggests that following the photolysis pulse one
should observe a loss of NO and a formation of NO2 equal to
the initial radical concentration. While this is true of the
quantity of NO2 formed, as illustrated by the transient IR
measurements shown in Figure 1, the loss of NO is nearly
double the initial peroxy concentration. Clearly, significant
secondary chemistry occurs on the timescale of reaction (5).
The principal contributor is
(99.999%), O2 (99.6%, hydrocarbons < 0.2 ppm), and ethane
(99.5%) are obtained from Michigan Airgas.
III. Results
C H O + NO + M f C H ONO + M
(7a)
2
5
2
5
a. Reaction Mechanism. Photolysis of a Cl2/C2H6/NO/O2/
N2 gas mixture produces ethylperoxy radicals via reactions (1)
and (2) within approximately 1 µs of the laser pulse. The
product study of Atkinson et al. shows that the reaction between
these radicals and NO proceeds to greater than 98% via the
channel
Further evidence for this secondary reaction is found from the
comparison of Figures 2 and 3. Figure 2 depicts the UV spectra
of C2H5O2 and C2H5ONO, as well as those of other species
that may contribute to the UV absorbance of the reaction
mixture. The time-resolved UV spectra of the reaction mixture
illustrated in Figure 3 reveal that, as ethylperoxy radicals are
lost, ethylnitrite is formed. This conclusion, of course, is
consistent with the previous product study of the ethylperoxy
8
C H O + NO f C H O + NO
2
(5)
2
5
2
2
5
8
and that less than 2% occurs by
C H O + NO + M f C H ONO + M
reaction with nitric oxide.
The formation of ethylnitrite is not necessarily the sole
secondary process to occur. In fact, the adduct formed by
2
5
2
2
5
2

1
2376 J. Phys. Chem., Vol. 100, No. 30, 1996
Maricq and Szente
the HNO channel to contribute (15 ( 10)% at 110 Torr of N2.
Note, however, that this issue is moot with regard to the IR
measurements of NO loss and NO2 formation, because a second
NO molecule is lost regardless of whether the ethoxy reaction
proceeds via (7a) or (7b).
A more minor secondary reaction is the one between
ethylperoxy radicals and the NO2 that is formed via reaction
(5);
C H O + NO + M f C H O NO + M
(8)
2
5
2
2
2
5
2
2
Its importance derives from the fact that the peroxyethylnitrate
product absorbs in the UV, as depicted in Figure 2B. The
reaction between ethoxy radicals and NO2,
C H O + NO + M f C H ONO + M
(9)
2
5
2
2
5
2
also likely produces a UV-absorbing product. However, the
importance of both reactions (8) and (9) diminishes as the NO
concentration increases. Since the NO2 concentration cannot
exceed [Cl]0, the ratio [NO]/[NO2] lies in the range of 5-10
under most of the experimental conditions; thus, the contribu-
tions of reactions (8) and (9) to the secondary chemistry should
remain below 10-20%.
For completeness we list three additional secondary reactions,
although these have a negligible affect on the derivation of k5:
Figure 2. Reference spectra of species that contribute to the time-
resolved UV absorbances for the C and
ONO represent the principal UV-absorbing species. HNO and
NO contribute weakly because they are minor products. NO
contribute weakly because their peak absorption cross sections
are about 9 times smaller than those of the other species.
H O
2 5 2
2 5 2
+ NO reaction. C H O
C
C
2
2
H
H
5
C H O + C H O f products
(10)
(11)
(12)
2
5
2
5
2
5
O
2
2
and NO
2
C H O + C H O f products
2
5
2
5
C H O + O f CH CHO + HO
2
2
5
2
3
The first two are not well studied; thus, their rate constants are
taken by analogy with the methylperoxy and methoxy coun-
terparts. Reaction (12) is one of atmospheric importance. Since
HO2 reacts with NO to form NO2 and OH, it could impact the
IR determinations of k5, which are based on NO loss and NO2
formation. Furthermore, the OH radical could regenerate
ethylperoxy radicals by reaction with C2H6 impacting both the
IR and UV measurements. For these reasons reaction (12) was
9
included in the model. Owing to the small value of k12 and
the low O2 concentrations of these experiments (with one
exception), only an insignificant amount of HO2 is formed over
the 100 µs timescales of these experiments. As evidenced by
the results, the impact of reaction (12) is not apparent even for
the one case in which a relatively high O2 concentration was
employed.
b. IR Measurements. Rate constants are determined in two
ways, as described in section II. For the IR data, such as
depicted in Figure 1, the concentration versus time dependences
of NO loss and NO2 formation are simultaneously fit to the
reaction model given in Table 1. As detailed above, the reaction
of ethylperoxy radicals with NO and the subsequent formation
of ethylnitrite are the principal reactions in this scheme. The
best fits are illustrated in Figure 1 for three initial NO
concentrations. Table 2 presents the measured rate constants
along with the experimental conditions.
In some experiments a systematic deviation between one or
both of the predicted concentration dependences and the
corresponding data traces is occasionally observed. This does
not necessarily indicate a poor fit. It is mainly due to small
systematic errors arising from assigning IR cross sections to
the NO and NO2 absorption lines that are probed. Were a fit
made to a single trace, as opposed to a pair of traces
simultaneously, a systematic error in IR cross section would
be counteracted by a “best fit” rate constant that deviates
2 2 6 2 2
Figure 3. Time-resolved UV spectra of a Cl /C H /NO/O /N gas
mixture at various times following photolysis of the mixture. The dotted
lines represent fits of eq 4 to the measured absorbances.
reaction (7) can dissociate and thereby provide an apparent
bimolecular pathway
C H O + NO f CH CHO + HNO
(7b)
2
5
3
for the reaction between ethoxy radicals and nitric oxide. This
channel is observed to contribute about 25% to the overall
5
reaction at low pressure, e.g., 2 Torr of He; however, its
contribution at total pressures of ∼100 Torr of N2 is uncertain.
For the analogous reaction between nitric oxide and methoxy
radicals, Frost and Smith10 find that in 3-125 Torr of argon
approximately 12% dissociate to give HNO. Since collisional
stabilization of the larger ethylnitrite adduct should be more
facile, the fraction k7b/k7 should be smaller than this value. This
is consistent with the present UV data (Vide infra) which show

Reaction between Ethylperoxy Radicals and Nitric Oxide
J. Phys. Chem., Vol. 100, No. 30, 1996 12377
TABLE 1: C
2
H
5
O
2
+ NO Reaction Mechanism
rate constant
reactiona
-
12 330/T
3 -1 b
5
6
. C2H5O2 + NO f C2H5O + NO2
a. C2H5O2 + C2H5O2 f
k5 ) 3.1 × 10
e
cm s
k6 ) 9.8 × 10-14e
ka/kb ) 10.5e-
-100/T
cm s
3 -1 1
2
C2H5O + O2
533/T
6
b. C2H5O2 + C2H5O2 f
C2H5OH + CH3CHO + O2
a. C2H5O + NO f C2H5ONO
. C2H5O2 + NO2 f C2H5O2NO2
. C2H5O + NO2 f C2H5ONO2
0. C2H5O + C2H5O2 f products
1. C2H5O + C2H5O f products
-
11
3 -1 9
7
8
9
1
1
1
k7 ) 4.4 × 10 cm s
-12 3
-1 1
-1 9
-1
k8 ) 8.6 × 10 cm s
-
11
3
k9 ) 2.8 × 10 cm s
-
12
3
k10 ) 2.6 × 10
cm s
-
11
3 -1
k11 ) 3 × 10
cm s
2. C2H5O + O2 f CH3CHO + HO2 k12 ) 6 × 10-14e
-550/T
cm s
3 -1 9
a
b
Reaction numbers correspond to those used in the text. Measured
in the present study.
TABLE 2: Ethylperoxy + Nitric Oxide Rate Constants
conditions
NO]0 [Cl]0
14
results
[
14
meas. (Torr) Ptot
O2 (×10 (×10
k5
-
3
-3
(× 10-12 cm s-1)
3
temp method C2H6 (Torr) (Torr) cm
)
cm
)
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
20
20
28
29
32
32
32
53
53
53
54
54
71
71
71
95
95
95
95
95
95
95
95
23
23
24
24
54
54
54
55
IR
IR
UV
UV
IR
IR
IR
IR
IR
UV
IR
IR
IR
IR
IR
IR
IR
IR
IR
IR
IR
IR
UV
IR
IR
IR
IR
IR
IR
IR
UV
2.3
2.3
3.2
3.1
2.7
2.7
2.7
2.5
2.5
2.6
2.3
2.3
2.7
2.7
2.7
3.0
3.0
2.7
2.7
3.2
3.2
2.5
3.1
2.6
2.6
2.8
2.8
3.4
3.4
3.4
3.7
98
98
11
11
98
96
20
20
20
31
31
33
11
11
21
21
21
32
32
11
11
15
15
26
26
26
26
12.7
12.7
36
36
36
57
110
52
144
31
64
120
32
61
63
48
97
28
58
110
32
65
43
86
28
59
87
58
21
39
47
92
30
57
106
51
8.4
15.7 ( 3.0
12.8 ( 2.4
12.9 ( 1.4
12.0 ( 1.9
10.2 ( 1.9
11.6 ( 2.2
10.7 ( 2.0
12.3 ( 2.3
11.2 ( 2.2
11.2 ( 1.5
12.3 ( 2.5
15.7 ( 2.8
10.7 ( 2.0
14.0 ( 2.7
13.7 ( 2.6
12.3 ( 2.5
10.3 ( 2.5
11.8 ( 2.2
9.3 ( 1.8
8.1 ( 1.5
8.6 ( 1.6
11.2 ( 2.1
8.5 ( 1.1
7.3 ( 1.5
11.0 ( 2.2
9.2 ( 1.7
8.4 ( 1.6
7.4 ( 1.4
6.9 ( 1.2
6.9 ( 1.2
6.9 ( 1.0
8.4
4.8
10.1
10
10
10
1.6
1.6
6.0
6.8
6.8
11.4
11.4
11.4
106
106
100
100
100
106
106
114
101
101
107
107
107
102
102
102
103
115
115
101
117
104
104
115
115
115
115
115
Figure 4. Concentration versus time profiles derived from UV spectra
such as those in Figure 3. The C H ONO, HNO, and C H O NO yields
2 5 2 5 2 2
have been combined into a single quantity and scaled by the factor
0.81. The C H O and product data were simultaneously fit to the
2
5
2
1.25
reaction model of Table 1, with k and the product concentration scaling
5
1.25
5.7
5.7
8.7
8.7
15.3
9.4
14
14
7.5
7.5
4.3
4.3
4.3
4.7
factor as adjustable parameters.
contribute <2% to the error. Assuming that the various
contributions to the overall error are statistically independent
leads to an overall error in k5 of approximately (19%.
c. UV Measurements. At a given time after photolysis,
the UV absorbance of the reaction mixture is, as given by eq 4,
a superposition of the contributions from each UV-absorbing
species present. Assuming that reactions (5) and (7) dominate
the chemistry, UV absorbances, such as illustrated in Figure 3,
would comprise of contributions from ethylperoxy radicals and
11
ethylnitrite. As time progresses, however, fits of the time-
121 112
resolved absorbances to reference spectra of these two species
begin to exhibit three shortcomings. First, the amount of C2H5-
ONO formed is only about 70% of the C2H5O2 lost. Second,
the fits of the spectra show small systematic deviations at
wavelengths shorter than about 215 nm. Third, the fits suggest
that the C2H5O2 concentration decays to 5-10% of its initial
value and not zero.
As is evident from Figure 2, both HNO and C2H5O2NO2
absorb strongly in the <215 nm range. Each is a possible
secondary product in the overall reaction scheme that is expected
to appear only in minor yield. Augmenting the fits with either
the C2H5O2NO2 or HNO reference spectrum, or both, signifi-
cantly improves the fits of the UV absorbances, particularly
below 215 nm. The dotted lines in Figure 3 illustrate this in
the case that both species are included in the fitting procedure.
This suggests that some mixture of C2H5O2NO2 and HNO
constitutes up to 30% of the secondary product yield. However,
the addition of these two species to the fitting procedure does
not eliminate the apparent residual C2H5O2 that is found. It is
possible that another, as yet unidentified, secondary species also
contributes to the long-time absorbance.
somewhat from the true value. Because of the overconstrained
nature of these fits, varying one rate constant cannot “correct”
for systematic errors in the cross sections of two distinct species
and small deviations are sometimes observed in the simultaneous
fits.
Aside from the fitting error, which is a few percent, systematic
uncertainties in the NO and NO2 IR cross sections, in the initial
C2H5O2 and NO concentrations, and in the rate constants
employed in the reaction model contribute to the overall error
in k5. The largest contributions come from uncertainties in
7
[
C2H5O2]0 and σNO. The UV cross section used to measure
the former quantity is accurate to ∼5%, and repeated measure-
ments of σNO indicate this quantity to have a similar uncertainty.
Each translates to an approximately 12% error in k5. Uncertain-
ties of 5% each in σNO2 and [NO]0 introduce smaller errors of
about 4% and 6% into k5. The best fit value of k5 is relatively
insensitive to the details of the secondary chemistry. An
uncertainty of 10% in k7 translates into an error of 3% in k5.
Twenty percent uncertainties in k8 and k9, as well as factor of
2
uncertainties in the rate constants for reactions (10)-(12),

1
2378 J. Phys. Chem., Vol. 100, No. 30, 1996
Maricq and Szente
3
% to the measurement of k5. The statistically combined error
is approximately 15%.
IV. Discussion
A fit of the measured rate constants, both UV and IR based,
+
1.5
1.0
to an Arrhenius expression yields the result k5 ) (3.1 ) ×
-
-
12 (330(110)/T
3 -1
10
e
cm s . A comparison of this expression with
the data is provided by Figure 5. The appearance of a small
negative temperature dependence is consistent with the reaction
proceeding via a rate-limiting association step to form C2H5O2-
NO followed by rearrangement and dissociation of the energized
complex.
There have been four previous determinations of k5 at room
2
temperatures. Adachi and Basco measured a rate constant of
-
12
3
-1
(
2.7 ( 0.2) × 10
cm s for the ethylperoxy reaction with
nitric oxide. This is about 3 times smaller than the value of
-12
3
-1
3
(
8.9 ( 3.0) × 10 cm s reported by Plumb et al. Adachi
and Basco used UV absorption at 250 nm to monitor ethylperoxy
loss, whereas Plumb et al. employed a flow system with mass
spectrometric detection. It is likely that absorption at 250 nm
by the ethylnitrite secondary product caused an apparent decay
of the ethylperoxy concentration that was slower than the actual
decay and, thereby, gave the unduly small rate constant. The
larger rate constant is supported by the recent measurements at
Figure 5. Variation of rate constant with temperature for the C
NO reaction. Error bars represent (2σ deviations, which include
fitting error and systematic uncertainties.
2 5 2
H O
-12
3
-1
4
2
95 K of (8.5 ( 1.2) × 10 cm s by Sehested et al., using
+
pulsed radiolysis and UV detection of the NO2 product, and
-12
3
-1
5
(
8.2 ( 1.6) × 10 cm s by Da e¨ le et al., using a discharge
flow technique. The present room temperature measurement
Because of the uncertainty in the nature of the last 30% of
the secondary reaction product, the measured yields of C2H5-
ONO, HNO, and C2H5O2NO2 are combined into a single
secondary product yield. Figure 4 illustrates the variations in
time of the C2H5O2 decay and the secondary product formation.
It is important to point out that none of the choices of which
secondary products to include when fitting the time-resolved
UV spectra result in more than a few percent change in the
determination of either the ethylperoxy or ethylnitrite concentra-
tions. They do affect the amount, but not time dependence, of
the total product yield. This is likely due to uncertainties in
one or more of the UV cross sections used to ascertain product
concentrations. Because of this, the C2H5O2 and the composite
secondary product concentration versus time profiles are fit to
the reaction model of Table 1 using two adjustable parameters:
k5 and a scaling factor for the absolute product concentration.
The second parameter assures that only the relative variation
of product concentration with time, but not its magnitude,
influences the best fit determination of k5.
-
11
3
-1
of k5 is (1.00 ( 0.15) × 10
cm s , a value that is slightly
higher than the previous determinations but well within the
combined error bars.
There are no previous measurements of the temperature
dependence of the C2H5O2 + NO rate constant with which to
compare our results. The recommended A factor and activation
1
energy for the analogous methylperoxy reaction with NO are
-
12
3 -1
4
.1 × 10 cm s and -180 K, respectively. Very recently,
Howard and co-workers have reported the somewhat steeper
temperature dependence of Ea ) -285 ( 60 K for the CH3O2
1
2
+
NO reaction as well as an activation energy of -360 ( 60
1
3
K for the i-C3H7O2 + NO reaction.
In each case a mild
negative temperature dependence is observed with a slope
comparable to the value of Ea ) -330 ( 110 K found in the
present study for the C2H5O2 + NO reaction.
Note Added in Proof: Concurrently with our study, Eberhard
and Howard (Int. J. Chem. Kinet.) have examined the temper-
ature dependence of the C2H5O2 + NO reaction using a low-
pressure flow tube and obtained rate constants nearly identical
to those reported here.
UV measurements of k5, along with the pertinent experimental
conditions, are recorded in Table 2 and displayed as a function
of temperature in Figure 5. The values of k5 determined by
the UV method are slightly smaller than the corresponding IR
measurements. This probably occurs because the small (ap-
parent) residual ethylperoxy concentration that is observed
causes an underestimation of k5. In spite of this, the two sets
of rate constants are consistent with each other within the
experimental uncertainties of the respective measurement
techniques. The errors in k5 are comprised of fitting error,
uncertainty in the initial NO concentration, and uncertainties
in the rate constants employed in the reaction model. Since
References and Notes
(
1) Lightfoot, P. D.; Cox, R. A.; Crowley, J. N.; Destriau, M.; Hayman,
G. D.; Jenkin, M. E.; Moortgat, G. K.; Zabel, F. Atmos. EnViron. 1992,
6A, 1805.
2
(2) Adachi, H.; Basco, N. Chem. Phys. Lett. 1979, 67, 324.
(3) Plumb, I. C.; Ryan, K. R.; Steven, J. R.; Mulcahy, M. F. R. Int. J.
Chem. Kinet. 1982, 14, 183.
(4) Sehested, J.; Nielsen, O. J.; Wallington, T. J. Chem. Phys. Lett.
1
993, 213, 457.
5) Da e¨ le, V.; Ray, A.; Vassalli, I.; Poulet, G.; Le Bras, G. Int. J. Chem.
Kinet. 1995, 27, 1121.
6) Maricq, M. M.; Szente, J. J.; Kaiser, E. W. J. Phys. Chem. 1993,
97, 7970.
(
[
Cl]0 and [C2H5O2]t are both based on the ethylperoxy reference
spectrum, the errors in these quantities originating from
uncertainty in σC H O are correlated and cancel. An uncertainty
(
2
5 2
of 5% in [NO]0 translates into an error of 6% in k5. An
uncertainty of 20% in k8 contributes an error of 3%. Uncertain-
ties by a factor of 2 in k10-k12 contribute errors of less than
(
(
7) Maricq, M. M.; Wallington, T. J. J. Phys. Chem. 1992, 96, 986.
8) Atkinson, R.; Aschmann, S. M.; Carter, W. P. L.; Winer, A. M.;
Pitts, J. N. J. Phys. Chem. 1982, 86, 4563.

Reaction between Ethylperoxy Radicals and Nitric Oxide
J. Phys. Chem., Vol. 100, No. 30, 1996 12379
(
9) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr, J.
A.; Troe, J. J. Phys. Chem. Ref. Data 1992, 21, 1125.
10) Frost, M. J.; Smith, I. W. M. J. Chem. Soc., Faraday Trans. 1990,
6, 1757.
(12) Villalta, P. W.; Huey, L. G.; Howard, C. J. J. Phys. Chem. 1995,
99, 12829.
13) Eberhard, J.; Villalta, P. W.; Howard, C. J. J. Phys. Chem. 1996,
00, 993.
(
(
8
1
(
11) Recall that by C2H5ONO we mean the composite C2H5ONO -
NO + NO2 spectrum.
2
JP9607935