Atmospheric fate of several alkyl nitrates: Part 2. UV absorption cross-sections and photodissociation quantum yields

Article abstract of DOI:10.1039/a701781b

The UV absorption cross-sections of methyl, ethyl and isopropyl nitrate between 233 and 340 nm have been measured using a diode array spectrometer in the temperature range 240-360 K. The absorption cross-sections of these alkyl nitrates decrease with increasing wavelength and decrease with decreasing temperature for λ > 280 nm. The photodissociation quantum yield for CH₃ONO₂ to produce NO₂ and CH₃O was found to be essentially unity at 248 nm using transient UV absorption methods. Production of O and H atoms in the photodissociation of methyl nitrate at 248 and 308 nm were found to be negligible using resonance fluorescence detection of the atoms. High quantum yields for O atoms were measured following 193 nm photolysis of methyl nitrate. The OH radical was measured to be a photoproduct with a very small quantum yield. Using the OH rate coefficients reported in the accompanying paper and the UV absorption cross-sections and the photodissociation quantum yields measured here, the first-order rate constants for atmospheric loss of methyl, ethyl and isopropyl nitrate were calculated. Photolysis was found to be the dominant atmospheric loss process for the three alkyl nitrates.

Full text of DOI:10.1039/a701781b

View Article Online / Journal Homepage / Table of Contents for this issue
Atmospheric fate of several alkyl nitrates
Part 2¤ UV absorption cross-sections and photodissociation quantum yields
Ranajit K. Talukdar, James B. Burkholder, Martin Hunter,” Mary K. Gilles, James M. Roberts
and A. R. Ravishankara*°
National Oceanic and Atmospheric Administration, Aeronomy L aboratory, 325 Broadway,
R/E/AL 2, Boulder, CO 80303 USA, and T he Cooperative Institute for Research in Environmental
Sciences, University of Colorado, Boulder, CO 80309, USA
The UV absorption cross-sections of methyl, ethyl and isopropyl nitrate between 233 and 340 nm have been measured using a
diode array spectrometer in the temperature range 240È360 K. The absorption cross-sections of these alkyl nitrates decrease with
increasing wavelength and decrease with decreasing temperature for j [ 280 nm. The photodissociation quantum yield for
CH ONO to produce NO and CH O was found to be essentially unity at 248 nm using transient UV absorption methods.
3
2
2
3
Production of O and H atoms in the photodissociation of methyl nitrate at 248 and 308 nm were found to be negligible using
resonance Ñuorescence detection of the atoms. High quantum yields for O atoms were measured following 193 nm photolysis of
methyl nitrate. The OH radical was measured to be a photoproduct with a very small quantum yield. Using the OH rate
coefficients reported in the accompanying paper and the UV absorption cross-sections and the photodissociation quantum yields
measured here, the Ðrst-order rate constants for atmospheric loss of methyl, ethyl and isopropyl nitrate were calculated. Photo-
lysis was found to be the dominant atmospheric loss process for the three alkyl nitrates.
The role of alkyl nitrates in the chemistry of EarthÏs lower
atmosphere has been described in Part 1.1 One of the impor-
tant parameters needed to assess the role of alkyl nitrates is
their atmospheric lifetime. The lifetime of alkyl nitrates may
be controlled by free-radical reactions, solar photolysis and
heterogeneous removal. In the preceding paper we described
our studies of the OH reactions with alkyl nitrates.1 In this
paper we report the absorption cross-sections and the photo-
dissociation quantum yields, the parameters needed to evalu-
ate the atmospheric photolysis rates of alkyl nitrates in
general and methyl nitrate in particular.
] CH ONO ] O(3P)
391
344
294
250
241
(1d)
(1e)
(1f)
(1g)
(1h)
3
] CH ] NO
3
3
] CH ONO ] H
2
2
] CH O ] NO ] O(3P)
3
3
] CH ONO ] O(1D)
Channel (1a) is possible at all wavelengths, i.e., the thermal
decomposition of CH ONO is exothermic to yield CH O
3
2
2
and HONO but hindered by a large barrier. Channel (1b) has
been identiÐed as a thermal decomposition process with an
activation energy of 40 kcal mol~1.2 Channel (1d) is spin-
forbidden, but has been suggested in ethyl nitrate photolysis
with a quantum yield of 0.14.9 Channels (1g) and (1h),
although not important in the atmosphere, may be important
for the interpretation of laboratory studies.
Alkyl nitrates exhibit broad unstructured UV absorption
spectra: there is one strong band centred around 200 nm with
weaker overlapping di†use bands that extend beyond 300
nm.2h4 The region of interest for photolysis in the troposphere
and lower stratosphere are wavelengths, j, of greater than 290
nm. The temperature-dependent absorption cross-sections for
methyl, CH ONO ,5,6 and ethyl nitrate, C H ONO ,7,8 in
In this work, we report the UV absorption cross-sections of
methyl, ethyl and isopropyl nitrate, (CH ) CHONO , as a
function of temperature, and the absolute photodissociation
quantum yields for methyl nitrate at 248 nm. The yields of H
and O atoms from the 193, 248 and 308 nm photolysis of
methyl and ethyl nitrates are also reported.
We combined the measured rate constants for the reactions
of OH with alkyl nitrates reported in the preceding paper with
the UV absorption cross-sections and the photodissociation
quantum yields reported in this paper to calculate the atmo-
spheric lifetimes of these compounds. The relative contribu-
tions of the di†erent loss processes to the atmospheric
removal of alkyl nitrates were also evaluated.
3 2
2
3
2
2 5
2
the wavelength range relevant to the atmosphere have been
reported. In calculating the lifetimes, the photodissociation
quantum yields have been assumed to be unity. Products of
the photolysis of ethyl nitrate8 at 308 nm have been shown to
be C H O and NO , independent of temperature and pres-
2
5
2
sure. The absolute quantum yields for the photodissociation
of CH ONO
measured directly.
3
2 under atmospheric conditions have not been
The energetically possible photolytic pathways for methyl
nitrate along with the wavelength thresholds, j
, for their
thres
occurrence are given below (all molecular species are assumed
to be in their ground electronic and vibrational state and con-
servation of spin or angular momenta are not considered):
Experimental
The apparatus, experimental procedures, and data analysis
used to measure the absorption cross-sections and quantum
yields have been extensively used in our laboratory and are
described in detail elsewhere.10h13 We present below only the
aspects needed to understand the present experiments.
j
/nm
thres
CH ONO ] CH O ] HONO
È
(1a)
(1b)
(1c)
3
2
2
]
]
CH O ] NO
697
497
3
2
CH O ] NO ] OH
2
¤
”
Part 1: ref. 1.
Present address: The Phillips Laboratory, Hanscom Air Force
UV absorption cross-sections
Base, MA, USA.
Also associated with the Department of Chemistry and Biochem-
istry, University of Colorado, Boulder CO 80309, USA.
°
Absorption cross-sections over the wavelength range 233È350
nm were measured between 240 and 360 K using a 30 W D2
J. Chem. Soc., Faraday T rans., 1997, 93(16), 2797È2805
2797

View Article Online
lamp coupled to a diode array spectrometer. The cross-
sections were measured in overlapping wavelength blocks of
beam. Spectra of the cell contents were recorded (0.7È3.5 s)
after photolysis and converted to spectral changes by using
the spectrum recorded immediately before the photolysis pulse
80 nm and then assembled to obtain the complete spectrum.
Absorption cross-sections at 298 K were determined by Ðlling
an absorption cell to a known pressure and measuring the
absorbance over a 80 nm block using the diode array spectro-
meter. At temperatures other than 298 K, a dual absorption
cell arrangement was used; the absorption spectra at various
temperatures were measured relative to that at room tem-
perature.10 The measured ratios of the cross-sections were
placed on an absolute scale using the 298 K cross-sections.
as the reference (I ). Thus, only changes in the concentrations
of the absorbers induced by the photolysis pulse were mea-
0
sured. In these experiments, the cell was Ðlled with a mixture
of CH ONO in a bath gas, photolysed by one laser pulse,
3
2
and pumped out after spectral acquisition. Signal averaging
was achieved by reÐlling the absorption cell with a fresh
sample for each photolysis pulse.
A tunable 662 nm diode laser was used for NO measure-
3
The absorption cross-sections of NO , an impurity in alkyl
ments.13 The diode laser beam with a cross-sectional area
2
nitrates, are much larger than those of the alkyl nitrates at
smaller than that of the photolysis beam was co-propagated
with the photolysis laser beam through the absorption cell
such that it traversed the centre of the photolysed volume.
The diode laser light was separated from the photolysis beam
and focused onto a photodiode whose signal was ampliÐed
(50]) and recorded in 5 ls wide bins. The detection limit for
j [ ca. 280 nm and their contributions to the measured
absorbance become very large as the wavelength increases;
e.g., it is 104 times larger at 340 nm. Hence, if present, NO
2
interferes with measurements of the alkyl nitrate cross-
sections. The ethyl nitrate sample contained a small amount
of NO2 impurity. Because of the small absorption cross-
sections of alkyl nitrates at j [ 290 nm, contributions due to
NO under these conditions was 5 ] 109 molecule cm~3 for
3
one shot (corresponding to an absorbance of ca. 1 ] 10~5).13
NO absorption were unavoidable in the case of ethyl nitrate.
A mixture of CH ONO in N was slowly Ñowed (linear Ñow
2
3
2
2
NO dimerizes to N O and, as the temperature is lowered,
velocity of ca. 20 cm s~1 at a total pressure of 110 Torr)
2
2 4
both NO and N O can contribute to the measured absorp-
tion. Therefore, a spectrum of NO in equilibrium with N O
was measured at each temperature using NO concentrations
approximately (within 90%) equal to those in the ethyl nitrate
through the cell. Concentrations of CH ONO in the absorp-
tion cell, measured using the diode array spectrometer, were
(1È5) ] 1015 cm~3.
2
2 4
3
2
2
2 4
2
The 248 nm laser Ñuence, f 248 (deÐned here as the number
of photons cm~2 pulse~1), at the entrance of the absorption
cell, was needed to calculate the absolute photodissociation
quantum yields from the measured concentrations of photo-
products. The Ñuence was determined by photolysing a
known concentration of N O and by measuring the concen-
samples present in the absorption cell. Structures in the
absorption spectra10 of NO and the absence of such struc-
2
ture in that of ethyl nitrate were exploited to determine the
contribution to the measured absorbance due to NO .
2
Methyl nitrate and isopropyl nitrate did not contain detect-
2
5
able levels of NO .
tration of NO generated using the relation:
2
3
In addition to the diode array spectrometer measurements
A
A662
B
1
between 240 and 360 nm, absorption cross-sections of alkyl
nitrates at 213.86 nm (Zn line) were determined at 298 K.
Known pressures of alkyl nitrates, measured using a 10 Torr
capacitance manometer, were Ñowed through a 2 cm long cell
and absorbances recorded. Alternatively, the cell was Ðlled
with an alkyl nitrate to a known pressure and the absorbance
measured. In addition, some measurements were carried out
using a 25 cm long cell and a 4% mixture of alkyl nitrate in
He.
f 248 \
NO3
(I)
p662 b@
p248 MU248 (NO )N[N O ]
NO3 N2O5
N2O5 N2O5
3
2 5
were b@ is the e†ective pathlength as described below. The
NO quantum yield, U248 (NO ), in the 248 nm photolysis of
3
5
N2O5
3
N O has been quoted by Ravishankara et al.14 to be unity,
2
(
0.96 ^ 0.06). However, the absorption cross-section of NO
3
at the peak of the 662 nm band they used was ca. 20% lower
than the currently accepted number.13,15 Hence, the NO
3
quantum yield in the 248 nm photolysis of N O from the
2
5
data of Ravishankara et al. should be ca. 0.78. Typical laser
Quantum yields for photodissociation of CH ONO
energies determined from N O actinometry were in the range
3
2
2 5
2
0È30 mJ cm~2 pulse~1.
Three experimental techniques were employed to quantify the
products of CH ONO photodissociation: (1) long-path tran-
Absolute quantum yields were calculated from the mea-
sured photoproduct absorbance using eqn. (II):
3
2
sient UV absorption spectroscopy was used to determine the
quantum yields of NO , NO , HONO, CH O and
A
Aj
B
1
U248(CH ONO ) \
x
x
(II)
2
3
3
CH ONO; (2) vacuum ultraviolet (VUV) resonance Ñuores-
3
2
pj b@ p248[CH ONO ] f 248
3
x
3
2
cence detection was used to measure the quantum yields of O
where U248(CH ONO ) is the quantum yield of product “xÏ in
and H atoms; and (3) laser induced Ñuorescence detection was
used to measure the yield of OH. The long-path absorption
was used to measure products of photodissociation in the 248
nm photolysis, while the atomic products in 193, 248 and 308
nm photolysis were quantiÐed using resonance Ñuorescence.
x
3
2
the photolysis of methyl nitrate at 248 nm; Aj is the measured
x
absorbance of “xÏ at wavelength j; pj is the absorption cross-
x
section of “xÏ at wavelength j; p248 is the absorption cross-
section of methyl nitrate at 248 nm; b@ is the e†ective
absorption pathlength through the cell. This correction term
is needed to account for the signiÐcant attenuation of the pho-
Long-path absorption. Most molecular photoproducts or
stable molecules produced from them via reactions with O or
NO were monitored using a diode array spectrometer
tolysis laser ([50%) by either N O
or CH ONO2
in the
2
5
3
absorption and is given by:
2
G
1 [ exp([b[y]p248)
H
equipped with a D2 lamp. NO3 was monitored using a 662
b@y \
y
(III)
nm diode laser via absorption.
[y]p248
y
The output of a 30 W D lamp was collimated, passed
2
where y \ N O or MN.
through
a
Pyrex absorption cell (pathlength \ 91 cm;
2 5
i.d. \ 2.2 cm), and focused onto the entrance slit of a 0.25 m
spectrometer equipped with a 1024 element diode array detec-
tor. The spectrometer covered a 150 nm segment with 0.5 nm
resolution. The photolysis beam (KrF excimer laser; 248 nm)
was co-propagated with the probe beam. The probe beam was
completely within the volume swept out by the photolysis
Atomic resonance Ñuorescence. The quantum yields of O(3P)
and H(2S) in the photolysis of methyl and ethyl nitrates were
determined by measuring the concentrations of the atoms
using time-resolved VUV atomic resonance Ñuorescence fol-
lowing the photolysis of alkyl nitrate with a known laser
Ñuence. The Ñuence was determined by actinometry where a
2798
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
known concentration of O (at 248 or 308 nm) or N O (at 193
similar with the absorption cross-sections decreasing with
increasing wavelengths. The absorption cross-sections increase
with increasing carbon chain length at all wavelengths mea-
sured. The variations in the cross-sections observed at the
longer wavelengths (j [ 320 nm) indicate the precision with
which we could measure small cross-sections.
Fig. 2 shows the ratios of absorption cross-sections for
methyl nitrate at temperature T to those at 298 K, p(j, T )/p(j,
298) as a function of wavelength. Between 235 and 280 nm,
the cross-sections are almost independent of temperature. At
wavelengths above 290 nm, the atmospherically important
region, the absorption cross-sections decrease with decreasing
temperature. There is nearly a factor of two decrease in the
cross-section at 330 nm in going from 298 to 240 K. There-
fore, for atmospheric photolysis rate calculations, it is impor-
tant to have an accurate knowledge of the absorption
cross-sections at low temperature in the long-wavelength
region.
3
2
nm) was photolysed to produce O or H [via reaction of O(1D)
with H ]. The apparatus and procedures used here are
2
described elsewhere.16 Quantum yields of O atoms were mea-
sured by adding ca. 10 Torr of N such that any O(1D) pro-
2
duced by photolysis is quenched to O(3P) before it reacted
with any other species. Thus, the measured O atom yields are
the sums of O(1D) and O(3P) yields. The detection sensitivities
of our apparatus for O and H atoms were 2 ] 109 and
2
] 108 cm~3, respectively, for a signal to noise ratio of unity
for 1 s integration.
The concentrations of the alkyl nitrates and N O in the
2
reactor were calculated from the measured Ñow rates (using
calibrated electronic mass Ñow meters) and pressure in the
photolysis cell. The ozone concentration was measured via
absorption of 253.7 nm (Hg lamp) light in a 100 cm long cell.
The absorption cross-sections of methyl and ethyl nitrates at
1
93 nm, 1.64 ] 10~17 and 1.57 ] 10~17 cm2 molecule~1, were
taken from Turberg et al.17 Cross-sections of alkyl nitrates at
other wavelengths were from the present work.
The temperature dependent absorption cross-section data
were parameterized using an expression used in previous
studies.5,6,10
Laser induced Ñuorescence of OH. The quantum yield for
the production of OH was measured using the apparatus
described in Part 1.1
C
p(j, T )
p(j, 298 K)
D
ln
\ B(j)(T [ 298)
(IV)
This parameterization reproduces the measured values to
within 1% for j \ 330 nm. Because the cross-sections dropped
o† very rapidly at T \ 240 K, we measured the cross-sections
at T [ 298 K to improve the accuracy of the B values. The B
parameters for each of the alkyl nitrates investigated here are
listed in Table 1. The B values for methyl, ethyl and isopropyl
nitrates as a function of wavelength, are shown in Fig. 3, along
Results and Discussion
In this section, we describe and discuss: (1) the UV absorption
cross-sections of methyl, ethyl and isopropyl nitrates as a
function of temperature, and (2) the quantum yields for
various products in the photolysis of methyl nitrate at 248 nm.
Quantum yields for H and O atoms and for OH radical in the
photolysis of methyl nitrate at a few wavelengths were also
measured. Our results are compared with those from earlier
investigations.
with those reported for HNO .10 The values vary with wave-
length in similar ways. This is not surprising since the
chromophore is the same, the nitrate group, in all these mol-
ecules.
3
Owing to the similarities in the wavelength dependence of
the B values for the three alkyl nitrates, it seems reasonable to
assume that the temperature-dependent absorption cross-
sections of other alkyl nitrates can be estimated from their 298
K absorption cross-sections.
The accuracy of our absorption cross-sections is primarily
determined by the reproducibility in the measurements (i.e.,
UV absorption cross-sections
The absorption spectra of methyl, ethyl and isopropyl nitrate
at 298 K measured here are shown in Fig. 1 and listed in
Table 1. The spectral shape for all three alkyl nitrates are
Fig. 1 Plots of absorption cross-sections of methyl, ethyl (]10) and
isopropyl (]100) nitrates at 298 K as functions of wavelength. Our
results are shown with cross-sections from previous studies shown for
comparison. MN: Rattigan et al. (open circle), Turberg et al. (Ðlled
square), Roberts and Fajer (open triangle); EN: Clemitshaw et al.
(
crossed circle), Turberg et al. (open square), Roberts and Fajer (Ðlled
Fig. 2 Ratio of the absorption cross-sections of methyl nitrate at
temperature T to those at 298 K as a function of wavelength for six
di†erent temperatures in the range 240È360 K
triangle); Clemitshaw et al. (open diamond), Turberg et al. (crossed
square), Roberts and Fajer (solid diamond).
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2799

View Article Online
Table 1 Absorption cross-sections (T \ 298 K) and the B-parameters (see text) for methyl, ethyl and isopropyl nitrate
methyl nitrate ethyl nitrate isopropyl nitrate
wavelength/
nm
p/10~20
B/10~3
p/10~20
B/10~3
p/10~20
B/10~3
236
238
240
242
244
246
248
250
252
254
256
258
260
262
264
266
268
270
272
274
276
278
280
282
284
286
288
290
292
294
296
298
300
302
304
306
308
310
312
314
316
318
320
322
324
326
328
330
332
334
336
338
340
342
344
346
348
350
352
354
356
358
360
8.27
6.77
5.66
4.88
4.34
3.97
3.72
3.56
3.44
3.34
3.25
3.16
3.07
2.96
2.84
2.72
2.57
2.42
2.27
2.10
1.94
1.77
1.61
1.45
3.35
3.44
3.49
3.45
3.36
3.24
3.10
2.97
2.88
2.82
2.79
2.80
2.81
2.84
2.89
2.94
3.01
3.08
3.15
3.24
3.32
3.40
3.51
3.61
3.72
3.84
3.96
4.08
4.21
4.38
4.56
4.72
4.91
5.14
5.43
5.67
6.00
6.27
6.68
7.13
7.54
8.00
8.88
9.32
10.3
10.7
11.5
13.8
15.5
16.1
11.9
9.68
8.00
6.77
5.91
5.31
4.90
4.63
4.45
4.31
4.20
4.09
3.98
3.86
3.73
3.58
3.41
3.23
3.05
2.85
2.64
2.43
2.22
2.02
1.82
1.62
1.43
1.25
3.10
3.18
3.28
3.30
3.28
3.20
3.09
2.98
2.88
2.81
2.77
2.77
2.78
2.83
2.88
2.93
3.01
3.09
3.16
3.26
3.36
3.46
3.57
3.68
3.80
3.94
4.08
4.20
4.35
4.55
4.73
4.89
5.10
5.36
5.61
5.89
6.23
6.55
6.91
7.28
7.98
8.51
9.09
9.65
10.5
11.1
11.8
13.0
13.5
14.3
15.0
15.7
17.3
16.2
13.2
10.8
9.03
7.72
6.79
6.12
5.69
5.40
5.21
5.07
4.94
4.83
4.70
4.56
4.41
4.23
4.04
3.84
3.61
3.38
3.15
2.90
2.66
2.42
2.18
1.94
1.72
1.52
1.33
1.12
0.961
0.820
0.690
0.575
0.475
0.389
0.314
0.246
0.195
0.154
0.119
0.0908
0.0689
0.0516
0.0384
0.0284
0.0209
0.0152
0.0111
0.008 10
0.005 82
0.004 19
0.002 94
0.002 10
0.001 58
0.001 09
0.000 799
0.000 514
0.000 354
0.000 233
0.000 211
0.000 181
2.65
2.74
2.76
2.76
2.70
2.62
2.55
2.49
2.47
2.47
2.48
2.54
2.60
2.67
2.76
2.84
2.93
3.05
3.14
3.25
3.38
3.49
3.60
3.76
3.93
4.04
4.21
4.42
4.60
4.76
4.95
5.18
5.37
5.65
5.90
6.28
6.63
6.98
7.26
7.83
8.45
9.03
9.38
9.68
10.5
11.7
12.2
11.9
13.3
15.1
13.8
12.1
13.9
14.5
17.2
1.29
1.14
1.00
0.871
0.754
0.642
0.519
0.435
0.364
0.298
0.241
0.194
0.154
0.121
0.0940
0.0723
0.0548
0.0412
0.0304
0.0228
0.0164
0.0118
0.008 68
0.006 18
0.004 47
0.003 11
0.002 16
0.001 61
0.001 16
0.000 812
0.000 468
1.10
0.947
0.784
0.665
0.561
0.465
0.382
0.301
0.242
0.191
0.151
0.117
0.0901
0.0686
0.0514
0.0381
0.0279
0.0202
0.0150
0.0107
0.007 64
0.005 47
0.003 82
0.002 70
0.002 04
0.001 98
0.001 02
0.000 591
0.000 415
0.000 225
precision), the uncertainties in measuring the pressure and
accounting for absorptions due to impurities. We estimate this
combined uncertainty to be ca. 5% at the 95% conÐdence
level at j O 320. At 340 nm, it is estimated to be 10È20%. We
expect that the cross-sections at j [ 340 nm and at T D 298
K, obtained by this formulation, will be reasonable because of
The absorption cross-sections of methyl nitrate reported by
previous investigators,7 with the exception of Turberg et al.,17
agree with our results. The data of Turberg et al. are higher
than ours at j [ 280 nm and the di†erence increases with j.
This discrepancy is likely to be due to their inability to
measure small absorptions.
(
a) the structureless nature of the absorption spectra and (b)
Our absorption cross-sections for ethyl nitrate at 298 K
agree very well with those of Turberg et al., Clemitshaw et
al.,7 and Zhu and Ding.8 The cross-sections reported by
the monotonic decrease in cross-sections with decreasing tem-
perature.
2800
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
CH O ] O ] CH O ] HO
(2)
3
2
2
2
CH O ] NO ] M ] CH ONO ] M
(3)
(4)
3
3
CH O ] NO ] CH O ] HNO
3
2
where k (298 K) \ 1.9 ] 10~15 cm3 molecule~1 s~1, k (298
2
3
4
K) \ 2.45 ] 10~11 cm3 molecule~1 s~1 at 1 atm and k (298
K) \ 4.8 ] 10~12 cm3 molecule~1 s~1. Following photolysis
of methyl nitrate in the presence of O or NO, spectra of the
2
products between 310 and 360 nm were measured. In this
wavelength range, CH O, CH ONO, NO and HONO, have
2
3
2
distinct absorption features. Thus, it was possible to determine
simultaneously the quantum yields for the production of
CH O, CH O, CH ONO and NO . The spectra obtained fol-
2
3
3
2
lowing CH ONO photolysis in excess O and NO are shown
3
2
2
in Fig. 4. These spectra were deconvoluted using the known
absorption spectra and cross-sections15 to quantify the yields
of CH ONO, HONO, CH O and NO . The small unstruc-
tured residual following spectral subtractions shows that we
have accounted for all the components.
3
2
2
Fig. 3 The plot of B-values, determined in this work, as a function of
wavelength for methyl (solid line), ethyl (dotted line) and isopropyl
Mixtures of 2% CH ONO in 600È700 Torr of O were
3 2 2
photolysed and the yields of CH O and NO were measured
(
broken line) nitrate. B values for HNO 10 are also shown for com-
3
2
2
parison. The B-values reported by Rattigan et al. (Ðlled circles) for
MN, Clemitshaw et al. (open circles) and Zhu and Ding (solid
triangles) for EN are shown for comparison.
to be 0.71 ^ 0.18 and 0.79 ^ 0.20, respectively. In this mixture,
reaction (2) competes with the radical recombination reac-
tion:15
CH O ] NO ] M ] CH ONO ] M
(5)
3
2
3
2
where k (298 K) \ 1.66 ] 10~11 cm3 molecule~1 s~1 at 1
Roberts and Fajer3 are systematically lower than ours at
j [ 290 nm, the di†erence increasing with wavelength.
Roberts and Fajer assumed that ethyl nitrate did not absorb
at j P 340 nm and that any absorption measured in that
5
atm. Under our conditions, 15% of the CH O radicals were
3
lost via reaction with NO produced in the photolysis. Conse-
2
quently, our measured yields of CH O and NO are lower
2
2
limits. Correcting for the loss of CH O due to reaction (5)
region was due only to NO . Hence, it is likely that they over-
3
2
leads to a photolysis quantum yield for channel (1b) of
corrected their measured absorbance at j \ 340 nm. The
0
.91 ^ 0.20. No absorption due to CH ONO or HONO was
agreement amongst reported cross-sections3,6,7,17 at 298 K
3
detected in the absence of added NO. Based on our detection
limits for CH ONO and HONO, we place upper limits for the
quantum yields for channels (1d) and (1h) and for (1a) of
U1d`1h
for isopropyl nitrate is good.
Rattigan et al.5,6 recently reported the temperature depen-
dence of absorption cross-sections for methyl nitrate. Clemit-
shaw et al.7 and Zhu and Ding8 measured the absorption
cross-sections of ethyl nitrate as a function of temperature.
Fig. 3 compares our B-values for both methyl and ethyl
nitrate with those from previous investigations. The agree-
ment among the various studies is good up to 320 nm, beyond
which the B-values tend to deviate. The larger di†erences at
longer wavelengths is indicative of the precision of the mea-
surements.
3
\ 0.1 and U \ 0.05.
1a
The products measured in excess NO yielded complemen-
tary results. Mixtures of ca. 1017 cm~3 of CH ONO and ca.
3
2
The 298 K absorption cross-sections at 213.86 nm mea-
sured using two di†erent cells were (1.67 ^ 0.02) ] 10~18 for
methyl,
(1.64 ^ 0.02) ] 10~18
for
deuteromethyl,
(
2.42 ^ 0.03) ] 10~18 for ethyl and (3.27 ^ 0.03) ] 10~18 cm2
molecule~1 for isopropyl nitrate. The inclusion of estimated
systematic errors at the 95% conÐdence level lead to the
following
methyl,
cross-sections:
(1.64 ^ 0.08) ] 10~18
(1.67 ^ 0.08) ] 10~18
for
for deuteromethyl,
(
2.42 ^ 0.12) ] 10~18 for ethyl and (3.27 ^ 0.17) ] 10~18 cm2
molecule~1 for isopropyl nitrates. These values were used to
determine the concentrations of the alkyl nitrates in the mea-
surements of OH reaction rate coefficients and the quantum
yields in the resonance Ñuorescence and pulsed laser induced
Ñuorescence systems.1
Photolysis quantum yields
Long-path absorption. Methyl nitrate was photolysed at 248
nm in the presence of radical scavengers, O and NO, to
2
convert the transient photolysis products to stable compounds
Fig. 4 Absorption spectra of products observed in the photolysis of
which could be detected via UV absorption by utilizing their
characteristic absorption features. In separate experiments, O
was used to convert CH O to CH O and NO was added to
convert CH O to CH ONO and CH O via the reactions (2)È
methyl nitrate: (a) in 600È700 Torr O (average of 20 photolysis
2
2
pulses) and (b) in the presence of NO (ca. 3 Torr) in 60È700 Torr of
N
(average of Ðve photolysis pulses). Reference spectra of the pro-
3
2
2
ducts used to determine the yields were obtained under pressure and
temperature conditions similar to those in the yield measurements.
3
3
2
(
4):15,18,19
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2801

View Article Online
1
017 cm~3 of NO in 60È700 Torr of N were photolysed and
of NO :15
2
3
the yields of CH O, HONO, CH ONO measured. Under
these conditions, the loss of CH O via reaction (5) was negligi-
ble compared to its loss via reactions (3) and (4). A typical
spectrum measured following photolysis is shown in Fig. 4(b).
The measured quantum yield was independent of pressure in
the range 60È700 Torr, with an average value of
2
3
NO ] O ] M ] NO ] M
(10)
2
3
3
where k (298 K) \ 2.57 ] 10~13 cm3 molecule~1 s~1 in 100
1
0
Torr of N , can be ruled out because O atoms were not
2
detected in 248 nm photolysis (see below). Therefore, our NO
quantum yield measurement provided a direct measure of the
3
U(CH ONO) \ 1.1 ^ 0.3.
yield of reaction (1e). The measured quantum yields are sum-
marized in Table 2.
3
Small amounts of CH O and HONO were also detected
2
following photolysis of CH ONO in an excess of NO
3
2
Oxygen and hydrogen atom resonance Ñuorescence
[
U(CH O) \ 0.3, U(HONO) \ 0.1]. The small yield of CH O
2
2
may be due to reaction (4); however, a small direct yield
Upon photolysis of (0.5È1) ] 1015 cm~3 of methyl and ethyl
nitrate at 248 and 308 nm, we did not detect any O atoms via
resonance Ñuorescence. Based on the detection sensitivity of
our apparatus, ca. 2 ] 109 cm~3 for 1 s integration, we place
upper limits for the oxygen atom quantum yields to be \0.01
and \0.03 for CH ONO and \0.007 and \0.018 for
[
U(CH O) \ 0.05] via channel (1a) or (1c) cannot be ruled out
2
from our data. It is possible that H atoms, channel (1f), gener-
ated CH ONO:
3
H ] CH ONO ] OH ] CH ONO
(6a)
(6b)
3
2
3
3 2
C H ONO at 248 and 308 nm, respectively (Table 3). The
]
H ] CH ONO
2
2
2
2 5
2
upper limits at 308 nm are larger because absorption cross-
sections of alkyl nitrates are smaller and, hence, smaller frac-
tions of the compounds were photolysed.
The rate constant for reaction (6) is k (298 K) \
6
1
.6 ] 10~13 cm3 molecule~1 s~1 (see below). Following
reaction (6a), OH could generate CH O, HONO and NO via
2
2
In contrast, when methyl and ethyl nitrates were photolysed
at 193 nm, signiÐcant amounts of O atoms were produced. By
comparing O atom signals from back-to-back experiments
the following sequence of reactions:
OH ] CH ONO ] H O ] CH ONO
(7)
(8)
(9)
3
2
2
2
2
where Ðrst N O and then alkyl nitrate was photolysed, the O
2
CH ONO ] M ] CH O ] NO ] M
atom quantum yields from methyl nitrate and ethyl nitrate
2
2
2
2
were determined to be 0.65 ^ 0.15 and 0.63 ^ 0.12, respec-
OH ] NO ] M ] HONO ] M
tively. The quantum yield for O(1D) from N O photolysis at
2
However, as with CH O, a part of our measured HONO yield
193 nm is known to be unity.22 In the presence of 10 Torr of
2
(
0.05) could be due to direct formation via channel (1a). Thus,
N , all the O(1D) was quenched to O(3P) within 0.3 ls. The
2
our data leads to a quantum yield for channel (1b) of U
\
absorption of resonance Ñuorescence signal by alkyl nitrates
1
b
U(CH ONO) ] U(CH O) of 1.3 ^ 0.3. The large uncertainty
were taken into account in calculating the quantum yield as
described elsewhere.11
As noted earlier, the O atom yields were measured in an
3
2
is due to uncertainties in the absorption cross-sections, laser
Ñuence and secondary product formation via side-reactions.
Photolysis pathway (1e) was studied by measuring time-
excess of N and, hence, the above values are the sum of
2
resolved absorptions at 662 nm, due to NO , after the pho-
quantum yields for O(1D) and O(3P). No attempt was made
3
tolysis of CH ONO ÈN mixtures at 248 nm. A small
to separately measure these yields. There was no observable
reaction of O(3P) with methyl nitrate or ethyl nitrate. The
upper limit of the rate coefficients for the reactions of O(3P)
with methyl and ethyl nitrates were calculated to be
\5.0 ] 10~15 cm3 molecule~1 s~1 at 298 K and are listed in
Table 4.
3
2
2
absorption signal (absorbance, A, \6 ] 10~5) was observed.
This signal corresponded to a quantum yield of U(NO ) \
3
0
4
.015 ^ 0.010. Varying the total pressure over the range
0È200 Torr had no e†ect on the yield. The absorption cross-
section of NO at 662 nm20,21 is over three orders of magni-
2
tude weaker than that of NO and, therefore, NO absorption
In the 248 nm photolysis of methyl nitrate, H atoms were
detected via atomic resonance Ñuorescence. We assign an
3
2
did not contribute to the measured signal. An indirect source
Table 2 Quantum yields for molecular products in the photolysis of methyl nitrate at 248 nm and 298 Ka
molecular quantum yields
bu†er gas/
experimental
conditions
excess O2
excess NO
OH
CH ONO
CH O
NO2
[0.79
HONO
\0.05
NO3
3
2
\0.1
[0.71
.9 ^ 0.2b
\0.3
0
1.17 ^ 0.29
0.96 ^ 0.24
0.1 ^ 0.03
N
\ 0.005
2
N
2
0.015 ^ 0.01
a All quoted errors are at the 95% conÐdence limit and include estimated systematic errors. b Corrected for the reaction of CH O with NO .
3
2
Table 3 Quantum yields of atomic products in the photolysis of methyl and ethyl nitrate at 193, 248 and 308 nma,b
reference
molecule/reaction
wavelength
/nm
molecule
species
quantum yield
CH ONO2
N O/193 nm
O(3P)
O(3P)
O(3P)
H(2S)
O(3P)
O(3P)
O(3P)
193
248
308
248
193
248
308
0.65 ^ 0.15
\0.01
3
2
O /248 nm
3
O /308 nm
\0.03
3
O /248 nm/H
0.05 ^ 0.02
0.63 ^ 0.12
\0.07
3
2
C H ONO
N O/193 nm
2
5
2
2
O /248 nm
3
O /308 nm
\0.02
3
a All quoted errors are at the 95% conÐdence limit and include estimated systematic errors. b VUV resonance Ñuorescence detection of H and O
atoms.
2802
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
Table 4 Rate coefficients of H atom and O atom reactions with
for the loss of H atoms (\k [CH ONO ]). The photolytically
6
3
2
alkyl nitrates at 298 K
generated OH, [OH] contributed ca. 10È15% to the total
0
OH that was produced. Based on the known yield of H atom,
bu†er gas/pressure
(Torr)
rate coefÐcient
/cm3 molecule~1 s~1
reaction
methoda
the quantum yield for OH from the direct photolysis of
methyl nitrate was calculated to be \0.005 [see Table 3]. The
O(3P) ] MN
O(3P) ] EN
H ] MN
N
N
\5 ] 10~15
\5 ] 10~15
RF
RF
LIF
RF
LIF
LIF
2
rate coefficient
k
at 298 K was calculated to be
6
2
(1.62 ^ 0.18) ] 10~13 cm3 molecule~1 s~1. The value of
He/SF /100È600
(1.63 ^ 0.18) ] 10~13
(3 ^ 1) ] 10~13
6
N /0 ] He/30
k (298 K) obtained from the analysis of the measured tempo-
2
7
D ] CD ONO
N /300
(2.3 ^ 0.7) ] 10~14
(9.3 ^ 0.7) ] 10~13
ral proÐles was (2.28 ^ 0.16) ] 10~14 cm3 molecule~1 s~1, in
3
2
H ] EN
excellent agreement with that reported in Part 1.1
a RF: VUV resonance Ñuorescence detection of H and O atoms. LIF:
pulsed photolysis laser induced Ñuorescence detection of OH/OD.
The OD radical was observed in the 248 nm photolysis of
CD ONO in N . The measured temporal proÐles of OD
3
2
2
were analysed as in the case of CH ONO photolysis
3
2
described above. The rate coefficient of OD production in the
reaction of with CD ONO , was measured to be
upper limit of 0.05 for the quantum yield for H atom pro-
duction in the 248 nm photolysis of CH ONO . The rate
D
3
2
3
2
(
2.3 ^ 0.7) ] 10~14 cm3 molecule~1 s~1 at 298 K and 100
constant for reaction (6) was obtained by analysing the H
atom temporal proÐles in the presence of varying concentra-
tion of CH ONO to be (3 ^ 1) ] 10~13 cm3 molecule~1 s~1.
Torr of N . The slower rate constant (a factor of ca. 7) of D
atom reaction with CD ONO compared to that of H atom
2
3
2
2
3
2
reaction with CH ONO is consistent with the majority of the
This value is similar to that obtained from the analysis of OH
temporal proÐles (see below).1 The reason for the large error
bar in the rate constant is the small quantum yield of H atom
and, the consequently, small H atom signal levels.
3
H ] CH ONO reaction proceeding via abstraction of hydro-
3
2
gen atom from the methyl group. The observed OH/OD pro-
duction was not due to reaction of H or D atoms with the
NO impurity since its level in the samples of methyl nitrate
The photodissociation quantum yields of O(3P) and H(2S)
determined using the VUV resonance Ñuorescence detection
are given in Table 2 along with the results obtained from
long-path measurements.
2
and deuterated methyl nitrate were undetectably small (see
Table 1 in Part 11 for the sample analyses).
Production of OH was observed when C H ONO was
2
5
2
photolysed at 248 nm in N ÈHe, as in the case of methyl
2
nitrate. Even though we did not directly detect H atoms,
OH quantum yields. During the course of the kinetic mea-
surements, we observed the production of OH following 248
nm photolysis of methyl nitrate. A typical OH temporal
based on analogy with methyl nitrate, we attribute OH gener-
ation to the production of H atoms in the primary photolysis
followed by reaction of H atom with C H ONO . The OH
proÐle measured in N bath gas is shown in Fig. 5. The
2
5
2
2
production rate coefficient determined from the temporal
proÐle of OH was (9.3 ^ 0.74) ] 10~13 cm3 molecule~1 s~1,
independent of the pressure of the bu†er gas and is listed in
Table 4.
instantaneous formation of OH implies a primary photolysis
product via channel (1c). The subsequent rise in the OH signal
results from the reactions of the photolysis products to form
OH. As shown above, H atoms are formed with a small yield
in the 248 nm photolysis and can react with CH ONO to
3
2
generate either OH or H via reaction (6). The CH ONO
2
2
2
Comparison with the previous results
radical, channel (6b), is expected to decompose thermally into
CH O ] NO and not to a†ect the OH signal. We attribute
The predominant pathway in the 248 nm photolysis of
2
2
the observed OH signal to reaction (6). The temporal proÐles
of OH were Ðtted to a biexponential expression of the form:
CH ONO at 298 K is channel (1b), CH O ] NO , at 298 K.
3
2
3
2
The yield for channel (1b) is P0.7. This yield also shows that
the photodissociation quantum yield for CH ONO is near
[
OH] \ a exp([bt) [ c exp([dt)
(V)
3
2
t
unity. The measured product yields were independent of pres-
sure in support of the unit dissociation quantum yield.
Because absorption at 248 nm appears to be due to the same
transition as those at longer wavelengths, we expect channel
In this expression, a [ c \ [OH] , the OH signal produced
0
by photolysis, b is the Ðrst-order rate constant for the loss of
OH (\k [CH ONO ]), and d is the Ðrst-order rate constant
7
3
2
(1b) to be the predominant photodissociation pathway at
j [ 248 nm and the quantum yield for dissociation to be
unity. A similar conclusion was reached for ethyl nitrate by
Zhu and Ding.8
Quantum yields of products in the photolysis of alkyl
nitrates have not been measured extensively in the wavelength
range of atmospheric interest, i.e., in the band centred around
280 nm. Zhu and Ding recently reported the quantum yields
of products in the 308 nm photolysis of ethyl nitrate over the
temperature range 278È298 K using cavity ring down spec-
troscopy. They obtained a quantum yield of 1.0 ^ 0.1 for
channel (1b), independent of temperature. Rebbert9 pho-
tolysed ethyl nitrate using a high-pressure mercury lamp and
attributed the observed Ðnal products (C H ONO, CH CHO,
2
3
5
3
2
NO and O ; and small amounts of CH ONO, CO, H , NO
2
2
and CO ) to three primary processes:
2
C H ONO ] hl ] CH CHO ] HONO
(11)
(12)
(13)
2
5
2
3
]
]
C H O ] NO
2
5
2
C H ONO ] O
2
5
Fig. 5 Temporal proÐle of OH observed in the photolysis of methyl
nitrate at 248 nm
By photolysing diethyl ketone at 378 K and assuming its dis-
sociation quantum yield to be 1.0, Rebbert calculated 0.094,
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2803

View Article Online
0.139 and 0.24 for the yields of processes (11), (12) and (13)
respectively. He also estimated the total photodissociation
yield of C H ONO to be ca. 0.5, which is a factor of ca. 2
2
5
2
lower than that reported by Zhu and Ding.8 Since Rebbert
used a broad-band photolysis, the values of the yields depend
critically on the overlap between the lamp spectrum and the
absorption spectra of ethyl nitrate and diethyl ketone. More-
over, it is uncertain whether the dissociation quantum yield of
diethyl ketone is unity at the photolysis wavelength and the
pressures he employed. For example, the photodissociation
quantum yield for acetone varies with pressure, wavelength
and temperatures.23,24
Gray and Rogers25 photolysed methyl nitrate with a low-
pressure mercury lamp as well as with a Kr-Ðlled Ñash lamp
and concluded that the primary photodissociation process
was channel (1b) but did not report an absolute quantum
yield. Gray and Style26 reported a quantum yield of 0.31 for
the photodissociation of ethyl nitrate at 253.7 and 265 nm;
C H O and NO were the products of photodissociation.
2
5
2
Luke et al.27 and Luke and Dickerson28 measured the
atmospheric photolysis rate of ethyl, n-propyl, n-butyl and
2-butyl nitrate in the wavelength range 290È700 nm as a func-
tion of temperature and solar zenith angle. They could
account for the observed products and photolysis rates by
assuming that only the alkoxy radical and NO were the
Fig. 6 Atmospheric loss rate constants for methyl, ethyl and isopro-
pyl nitrate as a function of altitude (i.e., temperature) due to reactions
with OH and photolysis. The J-value for ethyl nitrate calculated using
constant 298 K absorption cross-sections is also shown in the Ðgure
2
primary products in the photodissociation of alkyl nitrates
and their quantum yields were unity.
Renlund and Trott29 detected CH, NO and NO via che-
(
marked in the Ðgure as 298 K).
2
miluminescence following 193 nm photolysis of n-propyl
nitrate. They did not detect O atom formation in contrast to
our observation. No quantitative information was available
from their measurements.
Fig. 6. The J values calculated using just the 298 K absorption
cross-sections are higher than those obtained by including the
temperature dependence. For example at z [ 10 km it is a
factor of ca. 2 higher. Therefore, model calculations that did
not include the temperature dependence of the absorption
cross-sections have overestimated photolysis losses.
Yang et al.30 have recently studied the photodissociation of
jet-cooled CH ONO at 193 and 248 nm and have found
3
2
channel (1b) to be the principal channel with relative yields of
.7 and 1 at 193 and 248 nm, respectively. They also observed
0
production of O(1D) with a yield of 0.3 at 193 nm. Their result
for 248 nm photolysis agrees well with our observations.
Absolute photodissociation quantum yields, however, were
not measured in that work. Our measured O atom yield of
Luke et al.27 and Luke and Dickerson28 measured the rate
constants for solar photolysis of ethyl, n-propyl, n-butyl and
2-butyl nitrates as functions of solar zenith angle and tem-
perature. Their measured photolysis rate constant (1.1 ] 10~6
s~1) of ethyl nitrate at 288 K (for March 23, ca. 40¡ N, over-
head ozone column of 310 DU and solar zenith angle of 60¡)
is in good agreement with their and our calculated J values.
However, they reported that the J value decreased by a factor
of ca. 2 when the temperature dropped from 299 to 279 K.
This large change is inconsistent with our calculations, which
predict a decrease of ca. 20%.
The rate constants for the loss of alkyl nitrates via reaction
with OH, as shown in Fig. 6, are small fractions of the J
values. For example, reaction with OH contributes ca. 20% to
the total loss rate for ethyl and isopropyl nitrates and \10%
for methyl nitrate. However, if the rate constants for the reac-
tions of OH with alkyl nitrates measured by Kerr and
Stocker32 or Nielsen et al.33 are used, the contribution of OH
reactions would be signiÐcantly higher. Further, the rate con-
stants reported by Nielsen et al. increased as the temperature
decreased. For the same decrease in temperature, the UV
absorption cross-sections decreased. Thus, their rate constants
predict that the contribution of OH reactions would be
enhanced at the lower temperatures.
0
.65 ^ 0.15 at 193 nm, which is higher than that reported by
Yang et al., may be due to dissociation of vibrationally hot
NO . Such a dissociation was noted by Yang et al.
2
Atmospheric implications
The Ðrst-order rate constant for the loss of alkyl nitrate via
photolysis, J, is given by:
P
jmax
J \
I(j, z) p(T , j) U(T , j) dj
(VI)
jmin
where I(j, z) is the diurnally averaged solar Ñux at wavelength
j and altitude z, p(T , j) is the absorption cross-section at tem-
perature T and wavelength j, and U is the photodissociation
quantum yield. The diurnally averaged OH concentration
proÐle and the solar Ñux were calculated for 30¡ N for July 4
with an overhead ozone column abundance of 305 DU for a
US Standard Atmosphere.31 The temperature dependent
absorption cross-sections measured in this work were used
along with the standard temperature proÐle of the US Stan-
dard Atmosphere. The photodissociation quantum yield for
the alkyl nitrates was taken to be unity. The calculated J
values and the Ðrst-order rate coefficients for loss via reaction
with OH are shown in Fig. 6.
Clemitshaw et al.7 calculated J values and the rate coeffi-
cients for the loss via reactions with OH for various alkyl
nitrates using a two-dimensional model. Their calculated
noon-time J values as a function of altitude agree with ours.
They used previous literature values for OH reaction rate con-
stants with a negative temperature dependence and, hence,
concluded that OH reactions contribute signiÐcantly to the
loss of alkyl nitrates.
Inclusion of the temperature dependent absorption cross-
sections has a signiÐcant e†ect on the calculated photolysis
rate constants. The diurnally averaged J value for ethyl
nitrate, as a function of altitude, calculated using 298 K cross-
sections are compared with those calculated using the cross-
sections at temperatures characteristic of the given altitude in
As noted in Part 1, heterogeneous reactions of alkyl nitrates
in the troposphere are unlikely to be rapid. Reaction of alkyl
2804
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
1
1
1
2
3
4
R. F. Warren, T. Gierczak and A. R. Ravishankara, Chem. Phys.
L ett., 1991, 183, 403.
R. J. Yokelson, J. B. Burkholder, R. W. Fox, R. K. Talukdar and
A. R. Ravishankara, J. Phys. Chem., 1994, 98, 13144.
A. R. Ravishankara, P. H. Wine, C. A. Smith, P. E. Barbone and
A. Torabi, J. Geophys. Res., 1986, 91, 5355.
nitrates with other atmospheric oxidants are also not impor-
tant. Therefore, reactions of alkyl nitrates with OH radicals
and their photolysis are the two important loss processes and
the atmospheric lifetimes calculated here should account for
the removal of alkyl nitrates from the troposphere.
The atmosphere lifetimes of alkyl nitrates are long enough
15 W. B. DeMore, S. P. Sander, D. M. Golden, R. F. Hampson, M.
J. Kurylo, C. J. Howard, A. R. Ravishankara, C. E. Kolb and M.
J. Molina, Chemical Kinetics and Photochemical Data for use in
Stratospheric Modeling, JPL publication 94-36Ï, Jet Propulsion
Laboratory, Pasadena, CA, 1994.
to make them important NO reservoirs that can be trans-
x
ported. Therefore, the alkyl nitrates can be sources of NO and
NO in the remote troposphere. The atmospheric lifetimes of
2
alkyl nitrates vary with season as shown by Clemitshaw et al.
1
6
G. L. Vaghjiani, A. A. Turnipseed, R. F. Warren and A. R. Rav-
and, hence, their roles as reservoirs will depend on the season.
Because the OH abundance and the J values scale with the
available solar Ñux, the conclusions of Clemitshaw et al.
regarding the seasonal dependence of the lifetimes are robust
even though they used OH reaction rate coefficients that may
be too large.
ishankara, J. Chem. Phys., 1992, 96, 5878.
17 M. P. Turberg, D. M. Giolando, C. Tilt, T. Soper, S. Mason, M.
Davies, P. Klingensmith and G. A. Takacs, J. Photochem. Photo-
biol., A: Chem., 1990, 51, 281.
1
1
2
2
2
2
2
8
9
0
1
2
3
4
M. J. Frost and I. W. M. Smith, J. Chem. Soc., Faraday T rans.,
990, 86, 1751.
L. Batt, R. T. Milne and R. D. McCulloch, Int. J. Chem. Kinet.,
1
1
977, 9, 567.
A. Bongartz, J. Kames, F. Welter and U. Schurath, J. Phys.
Chem., 1991, 95, 1076.
The authors thank G. Frost and S. A. McKeen for help with
the J-value calculations. This work is supported in part by the
NOAA Climate and Global Change Research Program.
D. Hsu, D. L. Mont and R. N. Zare, Spectral Atlas of Nitrogen
Dioxide, 553 to 648 nm, Academic Press, New York, 1978.
M. Zeliko† and L. M. Aschenbrand, J. Chem. Phys., 1954, 22,
1
685.
References
H. Meyrahn, J. Pauly, W. Schneider and P. Warneck, J. Atmos.
Chem., 1986, 4, 277.
1
R. K. Talukdar, S. Herndon, J. B. Burkholder, J. M. Roberts and
S. A. McKeen, T. Gierczak, J. B. Burkholder, P. O. Wennberg, T.
F. Hanisco, E. R. Keim, R-S. Gao, S. C. Liu, A. R. Ravishankara
and D. W. Fahey, Geophys. Res. L ett., 1997, submitted for pub-
lication.
A. R. Ravishankara, J. Chem. Soc., Faraday T rans., 1997, 93,
2
787.
2
3
J. M. Roberts, Atmos. Environ. A, 1990, 24, 243.
J. M. Roberts and R. W. Fajer, Environ. Sci. T echnol., 1989, 23,
2
5
P. Gray and G. T. Rogers, J. Chem. Soc., Faraday T rans., 1954,
9
45.
5
0, 28.
4
5
L. E. Harris, J. Chem. Phys., 1973, 58, 5615.
2
2
6
7
J. A. Gray and D. W. G. Style, T rans. Faraday Soc., 1953, 49, 52.
W. T. Luke, R. R. Dickerson and L. J. Nunnermacker, J.
Geophys. Res., 1989, 94, 14.
O. Rattigan, E. Lutman, R. L. Jones, R. A. Cox, K. Clemitshaw
and J. Williams, J. Photochem. Photobiol., A: Chem., 1992, 66,
3
13.
2
2
3
8
9
0
W. T. Luke and R. R. Dickerson, Geophys. Res. L ett., 1988, 15,
1
6
7
O. Rattigan, E. Lutman, R. L. Jones, R. A. Cox, K. Clemitshaw
181.
and J. Williams, J. Photochem. Photobiol., A: Chem., 1992, 69,
A. M. Renlund and W. M. Trott, Chem. Phys. L ett., 1984, 107,
55.
X. Yang, P. Felder and J. R. Huber, J. Phys. Chem., 1993, 97,
0903.
1
25.
5
K. C. Clemitshaw, J. Williams, O. Rattigan, D. E. Shallcross, K.
S. Law and R. A. Cox, J. Photochem. Photobiol., A: Chem., 1997,
1
102, 117.
3
3
3
1
2
3
“U.S. Standard Atmospheres, 1976Ï, NOAA/NASA/USAF, 1976.
J. A. Kerr and D. W. Stocker, J. Atmos. Chem., 1986, 4, 253.
O. J. Nielsen, H. W. Sidebottom, M. Donlon and J. Treacy,
Chem. Phys. L ett., 1991, 178, 163.
8
9
0
L. Zhu and C-F. Ding, Chem. Phys. L ett., 1997, 265, 177.
R. E. Rebbert, J. Phys. Chem., 1963, 67, 1923.
1
1
J. B. Burkholder, R. K. Talukdar, A. R. Ravishankara and S.
Solomon, J. Geophys. Res., 1993, 98, 22937.
1
R. K. Talukdar, G. L. Vaghjiani and A. R. Ravishankara, J.
Chem. Phys., 1992, 96, 8194.
Paper 7/01781B; Received 13th March, 1997
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2805