Relative rate study of the reactions of acetylperoxy radicals with NO and NO2: peroxyacetyl nitrate formation under laboratory conditions related to the troposphere

Article abstract of DOI:10.1021/jp962266r

A relative rate study has been performed on the reactions CH₃C(O)O₂^· + NO₂ + M → CH₃C(O)O₂NO₂ + M (1) and CH₃C(O)O₂^· + NO → CH₃^· + CO₂ + NO₂ (2) in an atmospheric flow system in which the relative yields of CH₃C(O)O₂NO₂ (PAN) have been measured as a function of the ratio of reactants [NO]/[NO₂]. Over the temperature range 247-298 K, at a total pressure of approx.1 atm, the ratio was independent of temperature, k₁/k₂ = 0.41 ± 0.03, where the error limits are 2σ. The results are discussed with reference to other relative rate measurements of k₁/k₂ and to absolute measurements of k₁ and k₂. The atmospheric implications of the ratio k₁/k₂ in relation to PAN are briefly considered.

Full text of DOI:10.1021/jp962266r

J. Phys. Chem. A 1997, 101, 55-59
55
Relative Rate Study of the Reactions of Acetylperoxy Radicals with NO and NO2:
Peroxyacetyl Nitrate Formation under Laboratory Conditions Related to the Troposphere
Stephan Seefeld, David J. Kinnison, and J. Alistair Kerr*
EAWAG, Swiss Federal Institute for EnVironmental Science and Technology, ETH Zu¨rich,
CH-8600 Du¨bendorf, Switzerland
ReceiVed: July 26, 1996; In Final Form: September 30, 1996X
•
A relative rate study has been performed on the reactions CH3C(O)O2 + NO2 + M f CH3C(O)O2NO2 +
•
•
M (1) and CH3C(O)O2 + NO f CH3 + CO2 + NO2 (2) in an atmospheric flow system in which the relative
yields of CH3C(O)O2NO2 (PAN) have been measured as a function of the ratio of reactants [NO]/[NO2].
Over the temperature range 247-298 K, at a total pressure of ∼1 atm, the ratio was independent of temperature,
k1/k2 ) 0.41 ( 0.03, where the error limits are 2σ. The results are discussed with reference to other relative
rate measurements of k1/k2 and to absolute measurements of k1 and k2. The atmospheric implications of the
ratio k1/k2 in relation to PAN are briefly considered.
Introduction
spheric chemistry of PAN. Computer-modeling studies of the
troposphere have shown that the uncertainties in the rate
Peroxyacetyl nitrate (CH3C(O)OONO2, PAN) is formed in
the atmosphere in the oxidative degradation of many organic
compounds of both anthropogenic and biogenic origin.1,2 It is
an important oxidant component of summer smog and can cause
eye irritation and plant damage.3 Moreover, PAN acts as a
temporary reservoir for reactive intermediates involved in
summer smog formation, and it plays an important role in the
transport of NOx in the troposphere. It is therefore essential to
establish accurate data on the kinetics and mechanisms of its
formation and removal for inclusion in models of atmospheric
chemistry.
The precursor of PAN in the atmosphere is the acetyl peroxy
radical (CH3C(O)OO•), which is mainly formed from the
hydroxyl radical initiated oxidation of acetaldehyde. The initial
reaction produces an acetyl radical which is rapidly oxidized
to the corresponding acetylperoxy radical under atmospheric
conditions:
constants of the PAN chemistry are important when calculating
the ozone formation.4 The kinetics of the thermal decomposition
of PAN have been extensively studied and the rate coefficients
(k-1) are well established.5 Absolute rate coefficients have been
reported6-8 for reaction 1 and there have been two recent direct
measurements9,10 of k2 over a range of temperatures. The rate
coefficient ratio, k1/k2, has been measured in several relative
rate studies for PAN11-15 and a few analogous peroxyacyl
nitrates.16-19
Prior to the present study, however, only indirect determina-
tions of the ratio k1/k2 for PAN have been carried out. Cox et
al.11 and Cox and Roffey12 measured the NO to NO2 conversion
and absolute yields of PAN and derived k1/k2 from a kinetic
analysis involving complex reaction schemes. Hendry and
Kenley,13 Kirchner et al.,14 and Tuazon et al.15 investigated the
rate of the thermal decomposition of PAN as a function of the
[NO]/[NO2] ratio and obtained values of k1/k2 at temperatures
above 283 K.
HO• + CH3CHO f H2O + CH3CO•
In the present work, the ratio, k1/k2, was measured by a
method involving the production of acetylperoxy radicals from
the photolysis of biacetyl in the presence of O2:
CH3CO• + O2 + M f CH3C(O)OO• + M
In the polluted atmosphere, there are two main competing
reactions for the acetylperoxy radical:
(CH3CO)2 + hν f 2CH3CO•
(4)
(5)
CH3CO• + O2 + M f CH3C(O)OO• + M
CH3C(O)OO• + NO2 + M { k1 } CH3C(O)OONO2 + M
k-1
(1,-1)
By measuring the resulting relative PAN concentrations as a
function of the [NO]/[NO2] ratio, we have been able to measure
k1/k2 at atmospheric pressure and at lower temperatures than
hitherto reported.
k2
9
CH3C(O)OO• + NO
8 CH3C(O)O• + NO2
(2)
where reaction 2 is followed by the fast reaction
Experimental Section
fast
•
CH3C(O)O• 8 CH3 + CO2
(3)
Materials. The synthetic air (PanGas) was a mixture of 20%
O2 and 80% N2. Mixtures of nitric oxide (1004 ( 2 ppm) in
N2 and nitrogen dioxide (960 ( 2 ppm) in synthetic air were
supplied by Carbagas. The following chemicals were used
without further purification other than bulb-to-bulb distilla-
tion: biacetyl (2,3-butadione, Fluka, g99%), acetyl chloride
(Fluka, g99%), acetaldehyde (Fluka, g99.5%). PAN was
synthesized by the liquid-phase nitration of peracetic acid in
n-tridecane solution.20,21
Note that since PAN is relatively unstable, its decomposition
via reaction -1 has to be included.
The ratio of the rate constants, k1/k2, and the thermal
decomposition rate constant, k-1, are key values in the atmo-
* Author for correspondence, FAX: +44 121 472 8067; e-mail
A.KERR@bham.ac.uk.
X Abstract published in AdVance ACS Abstracts, November 15, 1996.
S1089-5639(96)02266-9 CCC: $14.00 © 1997 American Chemical Society

56 J. Phys. Chem. A, Vol. 101, No. 1, 1997
Seefeld et al.
consisting of a 1:1 mixture of propylene glycol and water was
circulated (thermocirculator, Lauda RKS 20) through the middle
chamber to maintain a uniform temperature in the reaction
vessel. Before entering the reaction vessel, the gas mixture was
equilibrated at the reactor temperature by passage through a
Pyrex coil surrounded by the thermostated fluid in the middle
chamber. A metal shield was wrapped around the outer surface
of the inner glass tube to prevent the irradiating light beam from
entering the middle chamber. The outermost chamber was a
vacuum jacket, to reduce the heat flux between the reactor and
its surroundings.
The reaction vessel was illuminated by a solar simulator
(Polytec XS 1000) consisting of a 1 kW Xe arc lamp. The
light passed to the flow reactor through a set of collimating
lenses, a water filter, an air mass filter (AM2, cutoff at 330
nm) and an NO2 filter. The latter was used to minimize
photolysis of NO2 in the reactor and consisted of a glass cell
with quartz windows filled with air containing about 1.5% NO2
at atmospheric pressure and had a transmission of 10% at 400
nm and 20% at 460 nm. Under these conditions biacetyl was
photolyzed over the spectral region 350-450 nm but photolysis
of NO2 was negligible.
Figure 1. Schematic diagram of the atmospheric flow reactor system.
Figure 2. Detailed drawing of the flow reactor.
Analysis. The PAN was measured using a gas chromato-
graph (Carlo Erba HRGC 5300) operated at 35 °C and fitted
with an electron capture detector (ECD) at 100 °C. Gas samples
were drawn from the exit of the reactor through a 2 m fused
silica capillary (i.d. 0.56 mm) at a flow rate of 20 cm3 min-1
and transferred via an automatic gas sampling valve using a
100 µL stainless steel loop into a cryotrap (Carlo Erba). The
compounds were separated on a 16 m capillary column24 (4%
PS-255) with H2 as the carrier gas at a flow rate of 10 cm3
Flow System. A schematic diagram of the apparatus is
shown in Figure 1. The experiments involved flowing an air
mixture through an irradiated reactor with a residence time of
about 50 s and analyzing the reactants and PAN product either
by gas chromatography (GC-ECD-FID) or with an NOx chemi-
luminescence analyzer. The experiments were carried out at
atmospheric pressure (965 ( 5 hPa) and over the temperature
range 247-343 K.
min-1
. The PAN retention time was 2.0 min under these
A steady flow of synthetic air containing biacetyl, NO, and
NO2 in the ppm range was established through the reaction
vessel. Mass flow controllers (MFC, Brooks, Model 5850E)
consisting of a flow sensor, a control valve and an electronic
control system allowed the separate control of all the concentra-
tions without affecting the total flow, which was constant over
the time scale of the experiments and usually at the rate of 400
cm3 min-1. The mass flow controllers were calibrated for air
(80% N2 and 20% O2) and were unaffected by the low
concentrations of reactants in the mixtures. These calibrations
were regularly checked using soap-bubble flow meters.
NO in nitrogen and NO2 in air were taken from commercial
cylinders. For biacetyl, an 8000 ppm mixture in synthetic air
was prepared and stored in a 25 dm3 glass bulb at a pressure of
about 1250 hPa. To prepare such a mixture, the evacuated bulb
was first filled with about 10 hPa of the biacetyl using a
precision pressure gauge and then synthetic air was added.
Preliminary attempts to generate the air-biacetyl mixture by
multiple dilution of a slow flow of air bubbled through a liquid
sample of biacetyl failed, owing to the lack of long-term
stability.
The combined gas streams from the MFCs were passed
through a gas-mixing device consisting of a glass tube containing
a series of Teflon baffles (Sulzer Chemtech) prior to entering
the flow reactor. All connections were made with Teflon or
PFA tubing (3 mm i.d.) which was covered with aluminum foil
to prevent prephotolysis of the reactants.
The flow reactor (see Figure 2) has already been used to
investigate gas-phase reactions under conditions related to the
troposphere.22,23 The Pyrex reaction vessel consisted of three
concentric chambers. The inner tube was the actual reaction
vessel with an internal diameter of 2.9 cm and a length of about
56 cm. The vessel was irradiated through evacuated glass
stoppers with quartz end windows. A thermostated fluid
conditions.
Linearity and calibration checks of the ECD were carried out
by measuring ppb levels of PAN simultaneously with the GC-
ECD, the NOx analyzer, and an FTIR spectrometer (BIORAD
FTS45). The PAN samples for the calibration were prepared
by passing air through a solution of PAN in n-tridecane and
subsequently diluting this gas stream with pure air using mass
flow controllers. The ECD response for PAN was linear up to
a mixing ratio of 500 ppb and the detection limit (3σ) was 0.5
ppb.
NO and NO2 were measured with a commercial chemilumi-
nescence NO-NO2-NOx analyzer (Thermo Environmental
Instruments Inc., Model 42) fitted with a stainless steel NO2 to
NO converter operating at a temperature of 625 °C. The
instrument was calibrated with standard mixtures of NO and
NO2. Additionally, the conversion rate (99%) was checked by
gas-phase titration of the NO standard with ozone.
Measurement of the Rate Data. During each experiment
only the concentrations of NO and NO2 were varied. All other
parameters such as temperature, total flow, reaction time, light
intensity, and the biacetyl concentration were kept constant. A
typical set of experimental results under the fixed conditions
described above is shown in Figure 3. Starting with no NO in
the mixture, the [NO]/[NO2] ratio was increased to a value of
about 4 and subsequently decreased stepwise back to 0, and at
each selected value of the [NO]/[NO2] ratio the relative amount
of PAN formation was measured. This procedure confirmed
that the system was stable over the time scale of the experiments.
Typical levels of reactants and products, in units of ppm, were
[NO] 0-4, [NO2] 1-5, [biacetyl] 25, and [PAN] 0.01-0.1.
As a check on the proposed mechanism for PAN formation
in this study, a range of additional experiments was carried out
as follows: (i) as described above but with different flow rates,

Peroxyacetyl Nitrate Formation
J. Phys. Chem. A, Vol. 101, No. 1, 1997 57
[PAN]t[NO])0 [NO]0 k
)
2 + 1
k1
[NO2]0
(V)
[PAN]t
On the basis of eq V the ratio of the rate constants k1/k2
was derived by a linear least-squares fit from the plots of
[PAN]t[NO])0/[PAN]t as a function of [NO]/[NO2]. A typical
plot of the present data is shown in Figure 4, which is derived
from the raw data of Figure 3.
The rate coefficient ratios found in this study are listed in
Table 1, where the errors quoted are 2σ. No significant
temperature dependence is found in the temperature range 247-
298 K, and the average of these measurements is 0.41 ( 0.03,
where the error is 2 times the standard deviation. Above 298
K the decomposition reaction -1 cannot be neglected, and these
measurements yield a ratio which is underestimated as will be
discussed later.
Figure 3. Plot of the measured relative mixing ratios of PAN (b) as
a function of [NO]/[NO2] (1) for T ) 25 °C. The dashed line shows
the PAN[NO])0 value used in the calculations.
153 and 916 cm3 min-1; (ii) with the photolysis of CH3COCl
as the radical source; (iii) with the photolysis of Cl2 in the
presence of CH3CHO as the radical source; (iv) with the reaction
vessel lined with a thin Teflon sleeve. In all cases the kinetic
data were consistent with the results of the main studies (see
Results and Discussion).
Discussion
Checks on the Proposed Mechanism of PAN Formation.
As noted in the Results and in Table 1, the measurements of
k1/k2 at 310 and 343 K show considerably lower values of the
ratio than over the temperature range 247-298 K. This effect
can be explained by the onset of the decomposition of PAN,
reaction -1, in relation to the time scale of the flow experiments.
Thus at 343 K we calculate the half-life of PAN at high [NO]/
[NO2] ratios to be 3.5 s compared to the residence time in the
flow system of ∼50 s for the standard flow rate of 400 cm3
min-1. As further evidence of this situation we have compared
the experimental results to those from a kinetic model of the
reaction scheme (reactions 1-5) using the rate coefficients
recommended by IUPAC.5 Details of the modeling calculations
are presented elsewhere,25 and the results are shown in Figure
5, where the agreement is reasonably satisfactory.
Further checks on the mechanism were carried out with
alternative sources of acetyl radicals, namely, the photolysis of
CH3COCl (with the air mass filter removed) and the photolysis
of Cl2 in the presence of CH3CHO. In both cases the results
for k1/k2 were consistent with the data derived from the biacetyl
photolyses. It should be noted, however, that when we used
relatively high ratios of [CH3CHO]:[Cl2], the values of k1/k2
increased to ∼0.6. This we believe was due to the presence of
substantial concentrations of OH radicals in the system, gener-
ated from the oxidative chain reaction12 involving CH3 radicals
produced as result of reaction 2. These OH radicals add to the
generation of PAN via OH-initiated attack on CH3CHO which
affects our assumption that the rate of formation of the acetyl
radical is independent of the [NO]/[NO2] ratio. Such additional
OH-initiated PAN formation is not possible with the photolytic
radical sources, CH3COCOCH3 and CH3COCl, since OH attack
on both of these molecules is unlikely to yield acetyl radicals
and hence acetylperoxy radicals and PAN under the conditions
of the present experiments.
Results
In the presence of relatively high concentrations of NO and
NO2, the formation of PAN and CO2 is governed by reactions
1 and 2. Thus we can write the rate equations:
d[PAN]/dt ) k1[CH3CO3]t[NO2]t
d[CO2]/dt ) k2[CH3CO3]t[NO]t
(I)
(II)
on the assumption that reaction -1 is negligible, which holds
at low temperatures. Taking the ratio of eqs I and II and
integrating over time assuming that [NO]0 and [NO2]0 are
constant give the following equation
[PAN]t k1 [NO2]0
)
(III)
k2
[CO2]t
[NO]0
where the subscripts 0 and t indicate concentrations at the
beginning of the experiment and at time t, respectively.
Initially we had intended to measure the CO2 formation as
well as the PAN formation. This, however, was not possible
as we discovered that the PAN decomposed to yield CO2 in
the infrared instrument which was used to measure the CO2
from the reaction. There were also additional problems in
dealing with the ambient levels of CO2 in the atmosphere of
the laboratory which interfered with the product CO2 levels in
the ppb range. We have therefore resorted to a kinetic analysis
based solely on the measurement of the PAN formation.
If the production of the acetyl peroxy radical is independent
of the [NO]/[NO2] ratio and there are no reactions of the acetyl
peroxy radical that are dependent on the [NO]/[NO2] ratio other
than (1) and (2), then the sum of ([PAN]t + [CO2]t) is
independent of the [NO]/[NO2] ratio. This sum can be
determined by measuring [PAN][tNO])0, which is [PAN]t when
[NO]/[NO2] ) 0 and therefore [CO2]t ) 0. Thus, independent
of the [NO]/[NO2] ratio, [CO2]t can be expressed as
Additional experiments were performed with the main radical
source from the photolysis of biacetyl but with higher and lower
flow rates through the reaction vessel, as recorded in Table 1.
The results show no systematic trend with increasing flow rate,
although in view of the higher errors involved at the lower flow
rate and the relatively low mixing ratio of NO and NO2 at the
higher flow rate, we have not included these data in calculating
the average value of k1/k2 over the temperature range 247-298
K.
[CO2]t ) [PAN][tNO])0 - [PAN]t
(IV)
Langer et al.26 have reported that heterogeneous reactions of
PAN are too slow to have an influence on PAN decomposition
under typical atmospheric conditions but may affect laboratory
Combination of eqs III and IV yields the expression

58 J. Phys. Chem. A, Vol. 101, No. 1, 1997
Seefeld et al.
Figure 4. Kinetic plot of the ratio of PAN measured as a function of
[NO]/[NO2] according to eq V for T ) 25 °C, based on measurements
shown in Figure 3.
Figure 6. Comparison of present data on k1/k2 with literature values:
(b) this study, (O) this study, at temperature where reaction -1 is
significant; (4) Cox et al.;11 (3) Cox and Roffey;12 (0) Kirchner et
al.;14 (]) Tuazon et al.;15 (- - -) Hendry and Kenley;13 (s) ratio of the
absolute rate coefficients of Bridier et al.6 and Villalta and Howard;9
(- - -) ratio of the absolute rate coefficients of Bridier et al.6 and
Maricq and Szente.10
TABLE 1: Rate Constant Ratios, k1/k2, as a Function of
Temperature at Atmospheric Pressure (965 hPa)
a
a
temp/K
k1/k2
temp/K
k1/k2
247
253
263
273
282
283
289
0.417 ( 0.019
0.391 ( 0.015
0.401 ( 0.015
0.436 ( 0.02b
0.428 ( 0.018
0.417 ( 0.02c
0.407 ( 0.01d
290
290
290
298
310
343
0.430 ( 0.025
0.444 ( 0.046e,g
0.468 ( 0.019f,g
0.405 ( 0.013
0.324 ( 0.015g
0.044 ( 0.005g
changes in the mixing ratios of NO2 and NO brought about by
these reactions are relatively small since the sum of the mixing
ratios of NO and NO2 was at least an order of magnitude higher
than the mixing ratios of the products, PAN and CO2. In
addition, the photolysis of NO2 was minimized by the presence
of the NO2 filter in the light train. Direct experimental
confirmation of the essential stability of the ratio [NO]/[NO2]
during the course of a reaction was obtained by switching on
and off the light or the flow of biacetyl while monitoring NO
and NO2.
a Errors are the sum of 2σ from linear least-squares fit and 2σ of
PAN[NO])0 determination. b Acetyl chloride as the radical source.
c Acetaldehyde + Cl2 as the radical source. d Teflon sleeve inserted
into the reactor. e Total flow 153 cm3 min-1. Total flow 916 cm3 min-1
.
f
g Not used to calculate the average (see text).
The photolysis of the PAN product during the course of the
experiments can be neglected for a number of reasons. First
there is very little overlap of the absorption spectrum27 of PAN
with the photolytic light (λ > 330 nm). In addition, during the
course of a run < 0.5% of the biacetyl was photolyzed, and the
absorption cross sections of biacetyl are orders of magnitude
higher than those of PAN under the experimental conditions.
Finally, if the PAN photolysis was significant it would be
expected that the values of k1/k2 would show a dependence on
contact time i.e., flow rate through the reactor, which is not the
case (see data of Table 1).
Comparison with Literature Data. Figure 6 shows a plot
of the present results in relation to previous determinations of
k1/k2 as a function of temperature and at pressures close to 1
atm. Our temperature independent value of k1/k2 ) 0.41 ( 0.03
is in agreement with the value k1/k2 ) 0.48 ( 0.07 recommended
in the IUPAC evaluation5 and by Wallington et al.28 on the basis
of the data of Cox and Roffey,12 Kirchner et al.,14 and Tuazon
et al.15 Note that in the present study we have been able to
extend the relative rate measurements down to a temperature
of 247 K and at the same time have considerably improved the
precision of the measurements.
Also shown in Figure 6 are values of k1/k2 calculated from
absolute measurements of k1 and k2. The values of k1 are from
Bridier et al.,6 which form the basis of the recommendations
for k1 in the IUPAC evaluation5 and the NASA panel evalua-
tion.29 These two groups differ in their representation of the
pressure effect on k1 according to the Troe formulation. Here
we have calculated k1 at 1 atm pressure as a function of
temperature from the values k0 ) 2.7 × 10-28(T/298)-7.1 [M]
cm3 molecule-1 s-1, k∞ ) 1.21 × 10-11(T/298)-0.9 cm3
molecule-1 s-1, Fc ) 0.30, and Nc ) 1.41 derived by Bridier et
al.6 over the temperature range 248-393 K. These k1 values
have been combined with the two recent determinations of k2
Figure 5. Comparison of present data on k1/k2 with a kinetic model
of the reaction scheme: (b) measured ratios, (solid line) modeled ratios.
measurements of PAN, through the decomposition of the
acetylperoxy radical or PAN itself at a surface. In the present
system, with a surface-to-volume ratio of about 1.4 cm-1 and
relatively high mixing ratios of NO and NO2, the possible
heterogeneous reactions26 are negligible in relation to the short
residence time (∼50 s) in the flow reactor. As further evidence
that heterogeneous reactions are not important in the present
system, an experiment was carried out with the Pyrex reaction
vessel lined with a Teflon sleeve. The kinetic data measured
with the Teflon surface were consistent with the results of the
main studies carried out with a Pyrex surface (see Table 1).
The above kinetic treatment assumes that [NO] and [NO2]
do not change significantly during the course of the reaction.
Of course, reactions 1 and 2 affect the mixing ratios of these
two compounds, and every methyl radical produced leads to
further NO to NO2 conversion.11 However, in our system the

Peroxyacetyl Nitrate Formation
J. Phys. Chem. A, Vol. 101, No. 1, 1997 59
to derive the lines plotted in Figure 6. Villalta and Howard9
have reported k2 ) 8.1 × 10-12 exp(270/T) cm3 molecule-1
s-1 over the temperature range 200-402 K, from a direct study
of the acetylperoxy radicals using mass spectrometry. Maricq
and Szente,10 on the other hand, have reported k2 ) 2.1 × 10-12
exp(570/T) cm3 molecule-1 s-1 over the temperature range 228-
353 K, from transient IR absorption of NO2 and time-resolved
UV spectroscopy of the acetylperoxy radicals. It is clear from
the two resulting lines shown in Figure 6 that there is a
significant difference, about 30% at 298 K, in the calculated
values of k1/k2 at atmospheric pressure based on the studies of
Villata and Howard9 and of Maricq and Szente.10 Until this
discrepancy is resolved, it is difficult to compare the calculated
values of k1/k2 with those measured in the relative rate studies,
although at face value the calculated ratios k1/k2 derived from
the study of Villata and Howard9 are in better agreement with
the present data than are those of Maricq and Szente.10
Atmospheric Implications. While at altitudes above 7 km
photolysis is the major loss process of PAN,30,27 the lifetime
(τ) of PAN in the warmer parts of the troposphere is controlled
by the rate of its thermal dissociation, reaction -1. In addition,
however τ is dependent upon reactions 1 and 2 and may be
expressed according to eq VI, based on a steady-state assumption
for the acetylperoxy radical concentration:12
(6) Bridier, I.; Caralp, F.; Loirat, H.; Lesclaux, R.; Veyret, B. J. Phys.
Chem. 1991, 95, 3594-3600.
(7) Addison, M. C.; Burrows, J. P.; Cox, R. A.; Patrick, R. Chem. Phys.
Lett. 1980, 73, 283-287.
(8) Basco, N.; Parmar, S. S. Int. J. Chem. Kinet. 1987, 19, 115-128.
(9) Villalta, P. W.; Howard, C. J. J. Phys. Chem. 1996, 100, 13624-
13628.
(10) Maricq, M. M.; Szente, J. J. J. Phys. Chem. 1996, 100, 12380-
12385.
(11) Cox, R. A.; Derwent, R. G.; Holt, P. M.; Kerr, J. A. Faraday Trans.
1 1976, 72, 2061-2075.
(12) Cox, R. A.; Roffey, M. J. EnViron. Sci. Technol. 1977, 11, 900-
906.
(13) Hendry, D. G.; Kenley, R. A. J. Am. Chem. Soc. 1977, 99, 3198-
3199.
(14) Kirchner, F.; Zabel, F.; Becker, K. H. Ber. Bunsen-Ges. Phys. Chem.
1990, 94, 1379-1382.
(15) Tuazon, E. C.; Carter, W. P. L.; Atkinson, R. J. Phys. Chem. 1991,
95, 2434-2437.
(16) Zabel, F.; Kirchner, F.; Becker, K. H. Int. J. Chem. Kinet. 1994,
26, 827-845.
(17) Kerr, J. A.; Stocker, D. W. J. Photochem. 1985, 28, 475-489.
(18) Kirchner, F.; Zabel, F.; Becker, K. H. Chem. Phys. Lett. 1992, 191,
169-174.
(19) Kenley, R. A.; Hendry, D. G. J. Am. Chem. Soc. 1982, 104, 220-
224.
(20) Nielsen, T.; Hansen, A. M.; Thomsen, E. L. Atmos. EnViron. 1982,
16, 2447-2450.
(21) Gaffney, J. S.; Fajer, R.; Senum, G. I. Atmos. EnViron. 1984, 18,
215-218.
(22) Harris, S. J.; Kerr, J. A. Int. J. Chem. Kinet. 1988, 20, 939-955.
(23) Semadeni, M.; Stocker, D. W.; Kerr, J. A. J. Atm. Chem. 1993,
16, 79-93.
(24) Grob, K.; Grob, G. J. High Resolution Chromat. Chromat. Commun.
1983, 6, 133-139.
(25) Seefeld, S. Laboratory Kinetic and Atmospheric Modelling Studies
of the Role of Peroxyacyl Nitrates in Tropospheric Photo-Oxidant Forma-
tion. Ph.D. Thesis, Swiss Federal Institute of Technology Zurich (ETH),
1997.
(26) Langer, S.; Wa¨ngberg, I.; Ljungstro¨m, E. Atmos. EnViron. 1992,
26A, 3089-3098.
(27) Talukdar, R. K.; Burkholder, J. B.; Schmoltner, A.-M.; Roberts, J.
M.; Wilson, R. R.; Ravishankara, A. R. J. Geophys. Res. 1995, 100, 14163-
14173.
(28) Wallington, T. J.; Dagaut, P.; Kurylo, M. J. Chem. ReV. 1992, 92,
667-710.
(29) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J.; NASA Panel for Data Evaluation; Chemical Kinetics and Photo-
chemical Data for Use in Stratospheric Modeling, EValuation Number 11,
JPL Publication 94-26, 1994.
k1 [NO2]
τ ) 1 +
(k-1)-1
(VI)
(
)
k2
[NO]
For example, at a temperature of 263 K in the free
troposphere, taking a typical31 [NO]/[NO2] ratio of 3, the lifetime
is 20 days, based on the value of k1/k2 determined here and the
value of k-1 recommended by IUPAC.5 This lifetime is 2.2
times lower when calculated simply on the basis of reaction
-1. Thus the ratio of rate coefficients, k1/k2, plays a significant
role in the lifetime of PAN in the troposphere and consequently
affects the transport of PAN and therefore the ozone generation.
References and Notes
(1) Stephens, E. R. AdV. EnViron. Sci. Technol. 1969, 1, 119-147.
(2) Roberts, J. M. Atmos. EnViron. 1990, 24A, 243-287.
(3) Taylor, O. C. J. Air. Pollu. Contr. Assoc. 1969, 19, 347-351.
(4) Yang, Y.-J.; Stockwell, W. R.; Milford, J. B. EnViron. Sci. Technol.
1995, 29, 1336-1345.
(30) Kleindienst, T. E. Res. Chem. Intermed. 1994, 20, 335-384.
(31) Carroll, M. A.; Hastie, D. R.; Ridley, B. A.; Rodgers, M. O.; Torres,
A. L.; Davis, D. D.; Bradshaw, J. D.; Sandholm, S. T.; Schiff, H. I.; Karecki,
D. R.; Harris, G. W.; Mackay, G. I.; Gregory, G. L.; Condon, E. P.; Trainer,
M.; Hubler, G.; Montzka, D. D.; Madronich, S.; Albritton, D. L.; Singh,
H. B.; Beck, S. M.; Shipham, M. C.; Bachmeier, A. S. J. Geophys. Res.
1990, 95, 10205-10233.
(5) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr, J.
A.; Troe, J. J. Phys. Chem. Ref. Data 1992, 21, 1125.