Kinetics of NO and NO2 evolution from illuminated frozen nitrate solutions

Article abstract of DOI:10.1021/jp055037q

The release of NO and NO₂ from frozen aqueous NaNO₃ irradiated at 313 nm was studied using time-resolved spectroscopic techniques. The kinetic behavior of NO and NO₂ signals during on-and-off illumination cycles confirms that NO₂ is a primary photoproduct evolving from the outermost ice layers and reveals that NO is a secondary species generated deeper in the ice, whence it eventually emerges due to its inertness and larger diffusivity. NO is shown to be more weakly held than NO₂ by ice in thermal desorption experiments on preirradiated samples. The partial control of gaseous emissions by mass transfer, and hence by the morphology and metamorphisms of polycrystalline ice, is established by (1) the nonmonotonic temperature dependence of NO and NO₂ signals upon stepwise warming under continuous illumination, (2) the fact that the NO, NO₂ or NO_x (NO_x ≡ NO + NO₂) amounts released in bright thermograms performed under various heating ramps fail to scale with photon dose, due to irreversible losses in the adsorbed state. Because present NO/NO₂ ratios are up to 10-fold smaller than those determined over sunlit snowpacks, we infer that the immediate precursors to NO mostly absorb at λ > λ_max (NO₃ ^-) - 302 nm.

Full text of DOI:10.1021/jp055037q

3578
J. Phys. Chem. A 2006, 110, 3578-3583
Kinetics of NO and NO2 Evolution from Illuminated Frozen Nitrate Solutions
C. S. Boxe,† A. J. Colussi,*,† M. R. Hoffmann,† I. M. Perez,‡ J. G. Murphy,‡ and R. C. Cohen‡
W. M. Keck Laboratories, California Institute of Technology, Pasadena, California 91125, and
Department of Chemistry and Department of Earth and Planetary Sciences, UniVersity of California,
Berkeley, California 94720
ReceiVed: September 6, 2005; In Final Form: January 24, 2006
The release of NO and NO2 from frozen aqueous NaNO3 irradiated at 313 nm was studied using time-
resolved spectroscopic techniques. The kinetic behavior of NO and NO2 signals during on-and-off illumination
cycles confirms that NO2 is a primary photoproduct evolving from the outermost ice layers and reveals that
NO is a secondary species generated deeper in the ice, whence it eventually emerges due to its inertness and
larger diffusivity. NO is shown to be more weakly held than NO2 by ice in thermal desorption experiments
on preirradiated samples. The partial control of gaseous emissions by mass transfer, and hence by the
morphology and metamorphisms of polycrystalline ice, is established by (1) the nonmonotonic temperature
dependence of NO and NO2 signals upon stepwise warming under continuous illumination, (2) the fact that
the NO, NO2 or NOx (NOx ≡ NO + NO2) amounts released in bright thermograms performed under various
heating ramps fail to scale with photon dose, due to irreversible losses in the adsorbed state. Because present
NO/NO2 ratios are up to 10-fold smaller than those determined over sunlit snowpacks, we infer that the
-
immediate precursors to NO mostly absorb at λ > λmax (NO3 ) ∼ 302 nm.
Introduction
in snow chamber experiments seem to be independent of
temperature between -30 and -20 °C,3 or positively correlated
with temperature in laboratory experiments on nitrate photolysis
in ice.9,10 The reasons for the larger HONO and HONO2 vertical
fluxes observed at the South Pole relative to other sites have
not yet been identified.6,8,11-13 A recent study suggested that
cation speciation underlies the significant differences between
inland and coastal Arctic regions for the release/uptake of these
acids.14,15
It is apparent that field measurements are affected by an open
set of physical and chemical factors that operate through still
unraveled mechanisms. Laboratory studies on snow chemistry
and photochemistry under controlled conditions may prove
essential for advancing our understanding of these phenomena.
We,10,15-17 and others,7,9 have performed quantitative studies
on the photochemistry of nitrate in its frozen aqueous solutions.
The monotonic temperature dependence of NO2, NO2- and NO
photoproduction rates above and below the normal freezing point
provide strong evidence that a similar mechanism operates in
fluid and frozen nitrate solutions down to subeutectic temper-
atures.10,16,17 We also showed that the amounts of NO2 photo-
desorbed during temperature-programmed nitrate photolysis
scale less than linearly with the duration of the experiments or
with nitrate concentration, and that NO2 emission rates display
abrupt transitions likely related to structural relaxations of the
ice matrix.15
The finding that elevated NOx (NOx ≡ NO + NO2) levels
develop in and above snowpacks during early spring1,2 has
revived interest in the nitrogen cycle over polar regions. NOx
emissions originate from the solar photolysis of embedded
nitrate, a ubiquitous contaminant of snow and ice absorbing
above 300 nm.3 The interpretation of NOx fluxes from snow
and ice measured at various geographical locations correspond-
ing to different sets of nitrate concentrations, solar irradiances/
zenith angles, and ambient temperatures, remains a challenge.
Thus, NOx fluxes, RNO , into the atmospheric boundary layer
x
over the South Pole, RNO ∼ 3 × 108 molecules cm-2 s-1 4
,
x
exceed those at other Antarctic or Arctic sites during the summer
solstice, even after factoring out differences in nitrate concentra-
tions and solar irradiances.5 If RNO ’s were exclusively deter-
x
mined by nitrate photolysis, the Summit (Greenland) site should
have recorded the highest values, at variance with observations.
Both NO and NO2 production rates (in molecules cm-3 s-1) in
the snowpack display significant and reproducible daily varia-
tions.2,6 NOx mixing ratios, which are proportional to NOx
production rates (or fluxes), are consistently larger inside the
snowpack than in ambient air during the day, pointing to a
photochemical process driven by sunlight. Field experiments
performed at noon over artificial snow made by freezing 100
µM NaNO3 solutions yield γ ) RNO/RNO ∼ 0.1, a value that
2
increases about 10-fold upon addition of free radical scavengers.7
Irradiation of Antarctic snow at λ g 295, 320, 345 and 385 nm
shows that NOx production ceases above 345 nm, confirming
nitrate as the primal chromophore, and that γ is an increasing
function of cutoff wavelength.3 The abnormally large NOx fluxes
detected at the South Pole have been tentatively ascribed to the
prevailing lower temperatures,8 although NOx production rates
Here we report time-resolved NO(g) and NO2(g) fluxes from
frozen nitrate solutions irradiated at λ ) 313 nm above -30
°C. The present study focuses on the differential responses of
R
NO and RNO to the onset and interruption of illumination, and
2
to temperature variations under isothermal or temperature-
programmed regimes.
Experimental Section
* Corresponding author. E-mail: ajcoluss@caltech.edu.
† California Institute of Technology.
‡ University of California.
The photoreactor used in these experiments has been de-
scribed in detail previously.15,16 About 6 mL of aqueous NaNO3
10.1021/jp055037q CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/23/2006

NO and NO2 Evolution from Frozen Nitrate Solutions
J. Phys. Chem. A, Vol. 110, No. 10, 2006 3579
(EM Science) solutions (pH e 6) are sprayed onto a cooled
finger (CF, a ) 304 cm2) where they instantly freeze into
porous, polycrystalline ice layers. The temperature of the ice
layers is controlled by a programmable cryogenic unit (Thermo
Neslab ULT-80) via coolant circulation through the CF. The
ice-covered CF is enclosed within a fused silica sheath (QS)
for illumination with the λ ) 313 ( 20 nm radiation emitted
by four Hg Pen-Ray UV lamps (UVP, model 90-0001-04),
which overlaps the λmax(NO3-) ) 302 nm absorption band of
nitrate. The overall irradiance incident on the ice layers: Ii )
3.0 × 1015 photons cm2 s-1, is determined using potassium
ferrioxalate as a chemical actinometer.18 The gaseous NO and
NO2 photoproducts released into the volume delimited by the
CF and the QS are continuously flushed into the detection zone
by means of zero-air carrier gas (FC ) 2.5 L min-1 at 1 atm,
293 K) within 11 s.
NO(g) is detected via NO + O3 f NO2* chemiluminescence
(Thermo Electron Corporation, Model 42C-TL NO-NO2-NOx
Analyzer).19,20 NO2* emits a broad continuum from 500 to 2800
nm (λmax ∼ 1300 nm),21 which is detected with a red-sensitive
(500-900 nm) photomultiplier tube. NO2(g) is detected by laser-
induced fluorescence. A pulsed dye laser is used to pump on
and off a pair of overlapping rotational lines of the A2B2 r
(000) X2A1 NO2 vibronic band at 585.26 nm. Fluorescence is
collected by a time-gated PMT detector in the range 750-1100
nm.22 Both detection systems were calibrated with certified
standard mixtures, have linear dynamic ranges exceeding 10
ppbv, i.e., much larger than the mixing ratios detected in present
experiments, and response times shorter than 2 s. The time
resolution of this setup is, therefore, largely determined by gas
transit times from the photolysis zone to the detectors. All the
results reported below represent the average of at least three
reproducible experiments.
Figure 1. NO2 emission rates, RNO , from a continuously illuminated
2
(at 313 ( 20 nm) 50 mM aqueous NaNO3 solution frozen at -30 °C,
and subsequently warmed along the temperature program indicated by
the solid line at the bottom of the plot. The dashed segments correspond
to estimated rates of photochemical NO2 production from the quantum
•
yields of OH formation in reaction 1, φ1, reported in ref 8.
NO and NO2 residence times in polycrystalline ice are antici-
pated to be different on account of their relative molecular sizes
and polarities. NO is a secondary species produced via reactions
•
3 and 4. Many species can act as sinks for the reactive OH
radicals. Their presence in natural snow and ice will critically
perturb the delineated photochemistry, which strictly applies to
frozen nitrate solutions.7
Under constant carrier gas flow rates, FC, the NO2 concentra-
tions detected in the gas phase are directly proportional to release
rates, rNO , from the irradiated solid:
2
æ1
Results and Discussion
r
) J
- ∂[NO2]S/∂t - C - D ) [NO2]FC
-æ1 + æ2
NO2
NO3
The photolysis of aqueous nitrate (ꢀmax ) 7.5 M-1 cm-1 at
302 nm) at λ > 300 nm, pH < 6 in fluid or frozen media
proceeds via reactions (1) (∼90%) and (2) (∼10%):9,10,23-29
(6)
Thus, rNO represents the balance between rates of photochemical
production, JNO æ1/(æ1 + æ2), accumulation in the solid, ∂-
2
-
hν
3
NO3- + H+ 98 NO2 + •OH
(1)
(2)
[NO2]S/∂t, and chemical, C, or photochemical destruction, D.
Because FC ) 41.7 cm3 s-1, the detection of 1 pptv NOx in the
gas phase (1 part in 1012 per volume ) 2.5 × 107 molecule
cm-3 at 1 atm, 293 K) in this setup is equivalent to the release
of r ) 1.0 × 109 molecules s-1 or, in terms of fluxes, of R )
3.4 × 106 molecules cm-2 s-1 from the illuminated area (see
Experimental Section). From the absorbance of ∼250 µm thick
frozen 50 mM nitrate layers: A ) 7.5 M-1 cm-1 × 5 × 10-2
M × 0.025 cm ) 9.4 × 10-3, the incident photon flux: Ii )
NO3- 9hν8 NO2- + O(3P)
The photolysis of both nitrite30-34 and NO2(g)35,36 yield NO:
hν
NO2- + H+ 98 NO + •OH
(3)
(4)
NO2 9
hν8 NO + O(3P)
3.0 × 1015 photons cm2 s-1, and the quantum yields of OH
•
production in reaction 1: φ1(•OH) ) 36.6 exp(-2400/T),8 we
estimate primary NO2 production fluxes: 3.0 × 1015 photons
cm2 s-1 × 9.4 × 10-3 × φ1(•OH) ) 5.6, 7.1, 11.6 and 14.2 ×
1010 molecule cm-2 s-1, at -30, -20, -10 and -4 °C,
respectively. These estimated NO2 production fluxes are shown
in Figure 1 as dashed lines, along with actual NO2(g) emission
NO2(g) increasingly absorbs at wavelengths longer than 313
nm, decomposing with high quantum yield: φ4 ∼ 1, at λ e
395 nm.35 The conjugated acid HONO (pKa ) 2.8)30 will
efficiently photolyze into (NO + •OH) in the gas phase.35 Note
that reaction 5 (k5 ∼ 2 × 1010 M-1 s-1):31
fluxes, RNO , measured over steadily illuminated 50 mM NaNO3
2
NO2- + •OH f NO2 + OH-
(5)
frozen layers. We do not attach much importance to the absolute
agreement between both sets of data because the estimated
fraction of incident light absorbed by the nitrate contained in
our ice layers does not take into account light scattering, layer
nonuniformity or the spectral mismatch between the actinometer
and nitrate absorption spectra. Effective emission fluxes, RNO
are expected, in principle, to be smaller than primary NO2
production fluxes due to potential NO2 losses prior to desorption.
-
•
rapidly converts NO2 into NO2, unless OH is efficiently
scavenged by other species.17
Thus, NO2 is a primary photoproduct of nitrate photodecom-
position, reaction 1, but can be also produced in secondary
reactions, such as reaction 5. The extent of NO2 photolytic
losses, reaction 4, is expected to increase with residence time.
,
2

3580 J. Phys. Chem. A, Vol. 110, No. 10, 2006
Boxe et al.
Figure 2. NO2 emission rates, RNO , from a frozen 50 mM aqueous
Figure 3. Arrhenius plots of RNO rates. 3: RNO ’s from Figures 1
2
x
2
NaNO3 solution illuminated during the isothermal stages, at -30, -20,
-10 and -4 °C, of the temperature program shown in Figure 1. The
inset zooms in the transients following the cessation of illumination at
the end of the -10 °C isothermal stage, and at the onset of illumination
early in the -4 °C stage.
and 2. 4: RNO’s from Figure 5.
The increasingly larger NO2 deficits observed at lower temper-
atures confirm that only a fraction of photogenerated NO2 is
released from ice. In contrast with the circumstantial agreement
between absolute rate values, the observed dependence of RNO
2
on temperature represents a real effect because it involves
relative rates obtained in a single experiment on the same ice
sample.
In the experiments of Figure 1, the temperature of the CF,
initially at -30 °C, was programmed to step up to -20, -10
and -4 °C at 0.7 °C min-1 after successive ∼ 1 h intervals.
RNO closely tracks temperature upon warming from -30 to -20
2
°C but slowly approaches a stationary value long after the system
has reached -10 °C, an effect that becomes more pronounced
Figure 4. NO emission rates, RNO, from a frozen 50 mM aqueous
NaNO3 solution illuminated during the isothermal stages, at -30, -20,
-10 and -4 °C, of the temperature program shown in Figure 1. The
inset zooms in the transients following the cessation of illumination at
the end of the -10 °C isothermal stage, and at the onset of illumination
early in the -4 °C stage.
at -4 °C. The slower, and nonmonotonic, response of RNO to
2
thermal perturbations at higher temperatures cannot be associ-
ated with an elementary photochemical process, but with
desorption from an increasingly accessible pool of preformed
NO2. Figure 2 displays the result of a related experiment in
which irradiation is interrupted during warming periods. In this
case, NO2 signals exponentially rise (decay) to steady-state
values with characteristic times τ ∼ 20 s, following the onset
(cessation) of illumination at lower temperatures, although they
still fail to reach steady state even after 1 h at -4 °C. The
observed τ’s are longer than the response of the detection system
and, therefore, represent a genuine effect. These experiments
assumption in these experiments that may, however, break down
for other ices.
Nitric oxide is released quite differently in these experiments
(Figures 4 and 5). Notice that the ratio of instantaneous rates:
γ ) RNO/RNO , varies between 0.1 and 0.043 (cf. Figures 1 and
2
5), depending on temperature. These values are about an order
of magnitude smaller than the γ values measured in the field.
-
The apparent discrepancy is likely due to the fact that both NO2
(Figures 1 and 2) confirm that RNO does not accelerate gradually
2
and NO2, the immediate precursors of NO, increasingly absorb
at wavelengths longer than 313 nm and, therefore, will undergo
more extensive photolysis under sunlight. The corollary is that
γ may also depend on factors affecting the solar spectrum such
as zenith angle, altitude, latitude, season and time of day.36 The
responses of NO fluxes into the gas phase to the onset of
illumination and darkness are qualitatively and quantitatively
different than in the case of NO2 (see the inset to Figure 4).
with T but markedly increases between -20 and -10 °C.
Furthermore, the dissimilar responses to temperature jumps
observed in Figure 1 vs Figure 2 starkly exposes the fact that
photochemical and thermal effects are not separable in these
systems. This phenomenon must be ascribed to the plasticity
of an ice matrix whose microstructure, as experienced by NOx
species, is determined not only by temperature but also by
external perturbations, such as ongoing photochemistry. A log
RNO vs 1000/T plot, based on the steady state RNO values of
RNO increases and decreases linearly, rather than exponentially,
as a function of time, with characteristic times τ ∼ 300 s vs
2
2
∼20 s for RNO . In the case of NO, τ is defined as τ ) 1/2 (tNOd
Figures 1 and 2, is shown in Figure 3. By forcing a linear
regression through these data, we obtain an apparent activation
2
- tNOdNO ). This behavior rules out NO as a primary
O
0
energy: ENO ) (38.3 ( 6.1) kJ mol-1, which is compatible
photochemical species. Furthermore, a linear RNO vs time
kinetics excludes desorption of a single monolayer of nonin-
teracting NO molecules but may be rationalized in terms of NO
emerging from increasingly populated deeper layers. The
2
with the value (E ) 41.8 kJ mol-1) we had reported before for
this process.10 An Arrhenius temperature dependence for RNO
2
is an acceptable, i.e., within the given error limits, working

NO and NO2 Evolution from Frozen Nitrate Solutions
J. Phys. Chem. A, Vol. 110, No. 10, 2006 3581
Figure 6. NO emission rates, RNO, from a 50 mM aqueous NaNO3
solution illuminated at -30 °C for 3 h, and then warmed at 0.5 °C
min-1 in the dark for 1 h.
Figure 5. NO emission rates, RNO, from a continuously illuminated
50 mM aqueous NaNO3 solution frozen at -30 °C, and subsequently
warmed along the temperature program shown in Figure 1.
plausibility of this scenario is supported by previous experiments
showing that NO2- is uniformly produced via reaction 2 down
to ∼300 µm in weakly absorbing nitrate-doped ice deposits,
whereas NO2, which is also formed along NO2- via reaction 1,
is only released from the outermost layers.10 Secondary NO2
photolysis, reaction 4, i.e., local NO production, is, therefore,
enhanced in deeper layers on account of NO2 longer residence
times, and possibly limited by the increasingly attenuated photon
fluxes. Note that NO2 residence times and light attenuation may
have, in principle, different depth profiles.
Also at variance with RNO , RNO markedly increases at -20
2
°C (cf. Figures 1 and 5). From the slope of the log RNO vs 1000/T
plot (Figure 3) we obtain an apparent activation energy: ENO
) (34.6 ( 6.5) kJ mol-1, which is similar to the ENO value
2
derived above. It should be emphasized again that, given the
nature of the observed phenomena, these apparent activation
energies reflect the interplay of various processes, rather than a
single elementary reaction. Some of these processes, such as
mass transport through a solid whose morphology undergoes
metamorphic transitions, may have non-Arrhenius temperature
Figure 7. NO emission rates, RNO, during a 0.5 °C min-1 dark
thermogram performed on a frozen 50 mM aqueous NaNO3 solution
that had been previously illuminated for 3 h (see Figure 6).
dependences. Note, however, that ENO is about twice as large
2
•
as E1, the activation energy for in situ production of OH
radicals, the complementary product of reaction 1, in ice.9 The
•
ratio RNO /R OH decreases at lower temperatures because the
2
competition between NO2 escape into the gas phase and
chemical and/or photochemical NO2 losses involves mesoscopic
displacements through powder ice, whereas the rates at which
•OH radicals escape the solvent cage reflect molecular displace-
ments.17,37 These arguments point to the inextricable coupling
between photochemistry and mass transfer in polycrystalline
ice. The coupling is inextricable because the competition
between photochemical change and mass transfer is not local,
as is the escape of the geminate products of reaction 1 from the
solvent cage, but a function of depth and the morphology of
the solid matrix.
To further decouple desorption from photogeneration, we
carried out experiments in which RNO and RNO were measured
2
over a frozen solution subjected to a two-stage schedule
consisting of (1) steady illumination at -30 °C for 3 h followed
by, (2) heating at 0.5 °C min-1 in the dark (Figures 6-9). In
stage 1 (Figure 6), RNO is found to slowly build up to a constant
value of ∼1.8 × 109 molecule cm-2 s-1 within ∼1 h. Early on
in stage 2 (Figures 6 and 7), RNO gradually falls off after
illumination has ceased. A broad desorption peak subsequently
ensues at ∼-11 °C, consistent with activated desorption/escape
Figure 8. NO2 emission rates, RNO , from a 50 mM aqueous NaNO3
solution illuminated at -30 °C for23 h, and then warmed at 0.5 °C
min-1 in the dark for 1 h.
from a distribution of occupied sites in a heterogeneous matrix
(Figure 7). In contrast, RNO rapidly rises up to ∼1.5 × 1010
2
molecule cm-2 s-1 upon illumination but continues to increase
up to RNO ∼ 2.4 × 1010 molecule cm-2 s-1 over 3 h (Figure
2

3582 J. Phys. Chem. A, Vol. 110, No. 10, 2006
Boxe et al.
Figure 11. RNO and RNO emission rates from a continuously il-
2
luminated (at 313 ( 20 nm) 50 mM aqueous NaNO3 solution frozen
Figure 9. NO2 emission rates, RNO , during a 0.5 °C min-1 dark
thermogram performed on a frozen 50 mM aqueous NaNO3 solution
at -30 °C and warmed to 5 °C at 0.1 °C min-1
.
2
from 7.6 at H ) 0.7 °C min-1 to 12.8 at H ) 0.1 °C min-1
The anomaly can be ascribed to the loss of photochemically
generated NO2 via thermal or photochemical reactions prior to
.
that had been previously illuminated for 3 h (see Figure 8; cf. Figure
6).
desorption. The condition RNO , RNO rules out secondary
2
photolysis, reaction 4, at the exclusive cause of NO2 deficits.
We have proposed that NO2 is hydrolyzed back to nitrate and
nitrite via a second-order reaction in [NO2], even at temperatures
below the NaNO3 eutectic:15
2NO2(g) + H2O(l) a
NO3 (aq) + NO2 (aq) + 2H+(aq) (7)
-
-
Given these considerations, the fact that ΣNO does not scale
with irradiation time may be also a consequence of NO2
hydrolysis, and the fact that the photolysis of aqueous nitrite is
less efficient that the photolysis of gaseous NO2, i.e., æ3 ,
æ4.24,35 Although reaction 7 is magnified in experiments
performed on more concentrated frozen nitrate solutions, we
found that NO2 hydrolysis is still significant in the micromolar
range.15 The reason is that solute concentrations in the micro-
scopic fluid phases in which these reactions actually occur are
vastly different from those measured upon melting. The fact
that the NO2 photogenerated in deep ice layers begins to desorb
above ∼-5 °C (see above) will only enhance the extent of
hydrolytic losses. Figures 10 and 11 dramatically show that time
and temperature are linked in these experiments via feedbacks
involving gas desorption, secondary thermal/photochemical
losses and polycrystalline ice morphology.
Figure 10. RNO and RNO emission rates from a continuously il-
2
luminated (at 313 ( 20 nm) 50 mM aqueous NaNO3 solution frozen
at -30 °C and warmed to 5 °C at 0.7 °C min-1
.
8). RNO precipitously drops 70-fold early in stage 2, before
2
bursting at ∼-5 °C (Figure 9). These experiments confirm that
NO is more loosely bound than NO2 to polycrystalline ice.38
The dissimilar diffusivities of NO and NO2 may be associated
with differences in molecular size and polarity in relation to
the microscopic structure of the ice matrix.39,40
Another perspective into the complexity of this system is
provided by the total amounts of NO, ΣNO, and NO2, ΣNO2,
liberated in the course of experiments performed by heating
frozen 50 mM NaNO3 solutions from -30 to 0 °C at H ) 0.7
and 0.1 °C min-1 (Figures 10 and 11) under steady illumination.
Implications for Polar Snowpack Chemistry
Our study shows that product distribution and transport within
ice critically affect the observable manifestations of photochem-
istry. These phenomena will depend on the microscopic and
mesoscopic structures of polycrystalline ice. However, there are
not canonical ices in nature.41 If there were, it would be still
difficult to reproduce them in the laboratory. Present experi-
mental conditions considerably differ from those prevailing in
sunlit snowpacks in a number of ways, such as (1) nitrate
concentration ranges (50 mM vs < 20 µM), (2) actinic photon
fluxes and spectra, and (3) ice morphology. As a result, absolute
NOx fluxes are about 2 orders of magnitude larger, and the γ )
As we have seen, instantaneous RNO and RNO do not represent
2
photochemical production fluxes because, among other reasons,
NO and NO2 might transitorily accumulate in the solid before
desorption (eq 6). However, if that were the only reason, all
the NO and NO2 produced by photolysis should be eventually
recovered after melting. Therefore, at constant irradiance,
because photon dose scales with irradiation time, i.e., with H-1
,
in the absence of secondary losses, both ΣNO and ΣNO2 should
increase 7-fold from the H ) 0.7 °C min-1 thermogram (Figure
10) to that at H ) 0.1 °C min-1 (Figure 11). This is clearly not
the case: ΣNO increases less than 2-fold, from 5.5 to 9.5,
whereas ΣNO2 only increases by a factor of 3, from 42 to 122
(Σ’s in arbitrary units), in these experiments. Thus, the longer
photolysis experiment fails to release the expected amounts of
NO and NO2 into the gas-phase, and â ) ΣNO2/ΣNO increases
R
NO/RNO ratio up to 10 times smaller than those over snow
2
fields. These considerations underline the fact that field experi-
ments, albeit perhaps more realistic, are site specific. Any
generalizations drawn on field experiments require prior iden-
tification of the relevant physical and chemical factors. Labora-

NO and NO2 Evolution from Frozen Nitrate Solutions
J. Phys. Chem. A, Vol. 110, No. 10, 2006 3583
(9) Chu, L.; Anastasio, C. J. Phys. Chem. A 2003, 107, 9594.
(10) Dubowski, Y.; Colussi, A. J.; Hoffmann, M. R. J. Phys. Chem. A
2001, 105, 4928.
(11) Yang, J.; Honrath, R. E.; Peterson, M. C.; Dibb, J. E.; Sumner, A.
L.; Shepson, P. B.; Frey, M.; Jacobi, H. W.; Swanson, A.; Blake, N. Atmos.
EnViron. 2002, 36, 2523.
(12) Honrath, R. E.; Lu, Y.; Peterson, M. C.; Dibb, J. E.; Arsenault, M.
A.; Cullen, N. J.; Steffen, K. Atmos. EnViron. 2002, 36, 2629.
(13) Dibb, J. E.; Huey, L. G.; Slusher, D. L.; Tanner, D. J. Atmos.
EnViron. 2004, 38, 5399.
(14) Beine, H. J.; Domine, F.; Ianniello, A.; Nardino, M.; Allegrini, I.;
Teinila, K.; Hillamo, R. Atmos. Chem. Phys. 2003, 3, 335.
(15) Boxe, C. S.; Colussi, A. J.; Hoffmann, M. R.; Murphy, J.;
Wooldridge, P. J.; Betram, T.; Cohen, R. C. J. Phys. Chem. A. 2005, 109,
8520.
tory results, obtained under better controlled conditions, may
identify relevant factors, anticipate possible effects, constrain
arguments, and eliminate implausible interpretations, but they
are not expected to provide quantitative predictions for specific
scenarios.
Thus, γ may be larger at lower temperatures because the onset
of enhanced ice permeability to NO occurs at lower temperatures
than for NO2. However, it is likely that permeability will depend
on ice morphology, i.e., on the mechanism of ice formation.
We confirm the prompter release of NO2 vs NO from nitrate-
doped ice upon illumination,6,42 a phenomenon arising from the
fact that NO2 is a primary photochemical product that only
desorbs from external ice layers, whereas NO is a secondary
species preferentially formed in deeper ice layers.
(16) Boxe, C. S.; Colussi, A. J.; Hoffmann, M. R.; Tan, D.; Mastro-
marino, J.; Case, A. T.; Sandholm, S. T.; Davis, D. D. J. Phys. Chem. A
2003, 107, 11409.
It has been pointed out that the consequences of nitrate
photolysis in the upper 10 cm of highly porous snow vs water
bodies are different because photolysis products are unlikely
to escape without further reaction from the latter.43 Our results
show that NO2 also undergoes hydrolysis before emerging from
ice, even below the eutectic. The significant increase of γ with
cutoff wavelength over Antarctic snow illuminated with light
from a Xenon-arc through long pass filters: γ ) 0.55 (no filter),
γ ) 0.82 (λ g 295 nm), and γ ) 2.78 (λ g 320 nm),3 indirectly
supports present low γ values at λ ) 313 ( 20 nm, confirms
that the immediate precursors of NO absorb at wavelengths
longer than nitrate itself, and suggests that γ may be a sensitive
function of the actinic light spectrum. Any possible role for
HONO as a photochemical precursor of NO will be conditional
to the local acidity of snow vs pKa(HONO) ) 2.8.30 Work is in
progress on the effect of radical scavengers and spectator ionic
species, such as those present in natural snow, on nitrate
photochemistry.
(17) Dubowski, Y.; Colussi, A. J.; Boxe, C.; Hoffmann, M. R. J. Phys.
Chem. A 2002, 106, 6967.
(18) Calvert, J. G.; Pitts, J. N. Photochemistry; Wiley: New York, 1966.
(19) Ridley, B. A.; Walega, J. G.; Dye, J. E.; Grahek, F. E. J. Geophys.
Res. 1994, 99, 25519.
(20) Ridley, B.; Walega, J.; Montzka, D.; Grahek, F.; Atlas, E.; Flocke,
F.; Stroud, V.; Deary, J.; Gallant, A.; Boudries, H.; Bottenheim, J.; Anlauf,
K.; Worthy, D.; Sumner, A. L.; Splawn, B.; Shepson, P. J. Atmos. Chem.
2000, 36, 1.
(21) Clough, P. N.; Thrush, B. A. Trans. Faraday Soc. 1967, 63, 915.
(22) Thornton, J. A.; Wooldridge, P. J.; Cohen, R. C. Anal. Chem. 2000,
72, 528.
(23) Mack, J.; Bolton, J. R. J. Photochem. Photobiol. A-Chem. 1999,
128, 1.
(24) Mark, G.; Korth, H. G.; Schuchmann, H. P.; vonSonntag, C. J.
Photochem. Photobiol. A-Chem. 1996, 101, 89.
(25) Daniels, M.; Meyers, R. V.; Belardo, E. V. J. Phys. Chem. 1968,
72, 389.
(26) Warneck, P.; Wurzinger, C. J. Phys. Chem. 1988, 92, 6278.
(27) Wagner, I.; Strehlow, H.; Busse, G. Z. Phys. Chem. Wiesbaden
1980, 123, 1.
(28) Zellner, R.; Exner, M.; Herrmann, H. J. Atmos. Chem. 1990, 10,
411.
(29) Deister, U.; Warneck, P.; Wurzinger, C. Ber. Bunsen-Ges. Phys.
Chem. 1990, 94, 594.
(30) Riordan, E.; Minogue, N.; Healy, D.; O’Driscoll, P.; Sodeau, J. R.
J. Phys. Chem. A 2005, 109, 779.
Acknowledgment. C.S.B. acknowledges support from the
Betty and Gordon Moore Foundation. This work was financed
by NSF grants ATM-0228140 (Caltech) and ATM-0138669
(Berkeley).
(31) Treinin, A.; Hayon, E. J. Am. Chem. Soc. 1970, 92, 5821.
(32) Zuo, Y. G.; Deng, Y. W. Chemosphere 1998, 36, 181.
(33) Rotlevi, E.; Treinin, A. J. Phys. Chem. 1965, 69, 2645.
(34) McEwen, K. L. J. Chem. Phys. 1961, 34, 547.
(35) DeMore, W. B.; et al. Chemical Kinetics and Photochemical Data
for Use in Stratospheric Modeling. Jet Propulsion Laboratory, 1997.
(36) Finlayson-Pitts, B. J.; Pitts, J. N. Atmospheric Chemistry; Wiley:
New York, 1986.
(37) Kim, J.; Keyes, T. J. Phys. Chem. B 2005, 109.
(38) Bartels-Rausch, T.; Eichler, B.; Zimmermann, P.; Gaggeler, H. W.;
Ammann, M. Atmos. Chem. Phys. 2002, 2, 235.
(39) Satoh, K.; Uchida, T.; Hondoh, T.; Mae, S. Proc. NIPR Symp. Polar
Metereol. Glaciol. 1996, 10, 73.
(40) Do, J. S.; Wu, K. J.; Tsai, M. L. Sens. Actuators 2002, B 86, 98.
(41) Hobbs, P. V. Ice physics; Clarendon Press: Oxford, U.K., 1974.
(42) Beine, H. J.; Domine, F.; Simpson, W.; Honrath, R. E.; Sparapani,
R.; Zhou, X. L.; King, M. Atmos. EnViron. 2002, 36, 2707.
(43) Wolff, E. W.; Jones, A. E.; Martin, T. J.; Grenfell, T. C. Geophys.
Res. Lett. 2002, 29, 1944.
References and Notes
(1) Honrath, R. E.; Peterson, M. C.; Guo, S.; Dibb, J. E.; Shepson, P.
B.; Campbell, B. Geophys. Res. Lett. 1999, 26, 695.
(2) Jones, A. E.; Weller, R.; Wolff, E. W.; Jacobi, H. W. Geophys.
Res. Lett. 2000, 27, 345.
(3) Cotter, E. S. N.; Jones, A. E.; Wolff, E. W.; Bauguitte, S. J. B. J.
Geophys. Res.-Atmos. 2003, 108.
(4) Oncley, S. P.; Buhr, M.; Lenschow, D.; Davis, D.; Semmer, S. R.
Atmos. EnViron. 2004, 38, 5389.
(5) Davis, D.; Chen, G.; Buhr, M.; Crawford, J.; Lenschow, D.; Lefer,
B.; Shetter, R.; Eisele, F.; Maulding, L.; Hogan, A. Atmos. EnViron. 2004,
38, 5375.
(6) Dibb, J. E.; Arsenault, M.; Peterson, M. C.; Honrath, R. E. Atmos.
EnViron. 2002, 36, 2501.
(7) Honrath, R. E.; Guo, S.; Peterson, M. C.; Dziobak, M. P.; Dibb, J.
E.; Arsenault, M. A. J. Geophys. Res. 2000, 105, 24183.
(8) Davis, D.; Nowak, J. B.; Chen, G.; Buhr, M.; Arimoto, R.; Hogan,
A.; Eisele, F.; Mauldin, L.; Tanner, D.; Shetter, R.; Lefer, B.; McMurry, P.
Geophys. Res. Lett. 2001, 28, 3625.