Photochemistry of adsorbed nitrate on aluminum oxide particle surfaces

Article abstract of DOI:10.1021/jp902252s

Nitrogen oxides, including nitrogen dioxide and nitric acid, react with mineral dust particles in the atmosphere to yield adsorbed nitrate. Although nitrate ion is a well-known chromophore in natural waters, little is known about the surface photochemistr

Full text of DOI:10.1021/jp902252s

7818
J. Phys. Chem. A 2009, 113, 7818–7825
Photochemistry of Adsorbed Nitrate on Aluminum Oxide Particle Surfaces
Gayan Rubasinghege and Vicki H. Grassian*
Department of Chemistry, UniVersity of Iowa, Iowa City, Iowa 52246
ReceiVed: March 13, 2009; ReVised Manuscript ReceiVed: May 11, 2009
Nitrogen oxides, including nitrogen dioxide and nitric acid, react with mineral dust particles in the atmosphere
to yield adsorbed nitrate. Although nitrate ion is a well-known chromophore in natural waters, little is known
about the surface photochemistry of nitrate adsorbed on mineral particles. In this study, nitrate adsorbed on
aluminum oxide, a model system for mineral dust aerosol, is irradiated with broadband light (λ > 300 nm)
as a function of relative humidity (RH) in the presence of molecular oxygen. Upon irradiation, the nitrate ion
readily undergoes photolysis to yield nitrogen-containing gas-phase products including NO2, NO, and N2O,
with NO being the major product. The relative ratio and product yields of these gas-phase products change
with RH, with N2O production being highest at the higher relative humidities. Furthermore, an efficient dark
reaction readily converts the major NO product into NO2 during post-irradiation. Photochemical processes
on mineral dust aerosol surfaces have the potential to impact the chemical balance of the atmosphere, yet
little is known about these processes. In this study, the impact that adsorbed nitrate photochemistry may have
on the renoxification of the atmosphere is discussed.
Introduction
groups in equilibrium with water vapor. It has been shown that
coadsorbed water molecules readily solvate adsorbed nitrate ions
on the aluminum oxide surface forming inner and outer sphere
complexes.7
In the atmosphere, surface-adsorbed nitrate on particulate
matter, for example, mineral dust and sea-salt aerosol, comes
from heterogeneous reaction of gas-phase nitrogen oxides
including NO2 and HNO3 with these particles.1,2 NOx (NO +
NO2) represents the major gas-phase nitrogen-containing an-
thropogenic pollutant, produced as a result of automobile and
industrial fuel combustion.3 In the lower atmosphere, these
nitrogen oxides participate in a complex series of chemical and
photochemical reactions to produce tropospheric ozone in a
nonlinear relationship with respect to the ozone concentrations.
Therefore, peak ozone levels are affected by the amount of NOx
present in the gas phase, which depends on the ground base
emissions and subsequent reactions in the lower atmosphere.4-6
Moreover, as noted above, atmospheric nitric acid, the main
reservoir species in the NOx cycle,5 can readily react with
particulate matter in the atmosphere to yield adsorbed nitrate.4
Therefore, understanding all processes that impact NOx and
reservoir species such as HNO3 is important in atmospheric
chemistry and air quality.
Oxide minerals represent an important and reactive compo-
nent of mineral dust aerosol. Given the fact that aluminum
oxides and aluminum silicates contribute ∼8% by mass to the
total dust burden in the atmosphere,7-9 aluminum oxides can
be used as model systems to begin to understand various aspects
of nitrogen oxide reactions on mineral dust aerosol. Several
experimental and theoretical studies have shown that hetero-
geneous uptake of nitric acid on aluminum oxide and alumi-
nosilicates yields a chemisorbed nitrate layer on the particle
surface.4,7,10-12 Vibrational frequencies of adsorbed nitrate have
been assigned and interpreted using quantum chemical calcula-
tions of binuclear alumina cluster models.7 Under humid
conditions, atmospheric water vapor readily adsorbs to oxide
particles and dissociates, resulting in a hydroxyl-terminated
surface.13 Adsorbed water can hydrogen bond to these hydroxyl
Although nitric acid adsorption and water uptake on alumi-
num oxide surfaces have been previously studied, much less is
known about the photochemistry of adsorbed nitrate on these
particle surfaces. In natural waters, nitrate is an important
chromophore14,15 with an absorption in the solar region of the
spectrum at 302 nm (ε ) 7.2 M-1 cm-1) attributed to the n-to-
π* electronic transition. Excitation of this transition results in
the photochemical production of nitrogen dioxide and nitrite
ion under acidic conditions.16-19 This study is a follow up to a
previously published communication20 that showed that the
photochemistry of adsorbed nitrate resulted in the formation of
several gas-phase products including N2O, NO, and NO2.
Besides the important ramifications of this photochemical
reaction converting nitrate into NOx and N2O, another interesting
point of the earlier study is that the product distribution is
dependent on the ambient conditions, that is, RH and the
presence of molecular oxygen. Here we expand on this initial
investigation and more thoroughly investigate the relative
humidity (RH) dependence of the reaction products and kinetics.
In addition, we present a study of a surface-mediated dark
reaction converting NO to NO2 during post-irradiation.
Experimental Section
Sources of the Oxide Powder and Other Reagents.
Commercially available γ-Al2O3 (Degussa, aluminum oxide C)
with a surface area of 101 ( 4 m2 g-1 was used in these
experiments. The surface area was determined using a Quan-
tachrome Nova 4200e multipoint BET apparatus. Dry gaseous
nitric acid was taken from the vapor of a 1:3 mixture of
concentrated HNO3 (70.6% HNO3, Mallinckrodt) and 95.9%
H2SO4, (Mallinckrodt). Distilled H2O (Fisher, Optima grade)
was degassed prior to use.
* To whom correspondence should be addressed. E-mail:
vicki-grassian@uiowa.edu.
10.1021/jp902252s CCC: $40.75  2009 American Chemical Society
Published on Web 06/17/2009

Photochemistry of Adsorbed Nitrate
J. Phys. Chem. A, Vol. 113, No. 27, 2009 7819
Figure 1. FTIR spectra of surface-adsorbed nitrate under different relative humidity conditions (RH < 1, 20, 45, and 80%) and 298 K (a) before
broadband irradiation (t ) 0 min) and (b) after 420 min of irradiation (t ) 420 min) (solid line). Each spectrum was referenced to the clean oxide
spectrum prior to exposure to nitric acid. Gas-phase absorptions were then subtracted from each spectrum. The dashed-line spectra shown in part
b are the difference spectra obtained by subtracting the spectral data at time t ) 420 min from those at t ) 0 min. Some of the more notable
absorptions that decrease and increase in intensity with irradiation are marked in part b.
Transmission Fourier Transform Infrared Spectroscopy.
Photochemistry of adsorbed nitrate was monitored via transmis-
sion FTIR spectroscopy. The infrared cell consists of a stainless
steel cube with two barium fluoride windows and a sample
holder. The inside of the stainless steel cube is coated with
Teflon to avoid HNO3 decomposition on the walls of the infrared
cell. For these measurements, the oxide powder (5.7 ( 0.4 mg)
was prepared by pressing onto half of a tungsten grid and the
other half of the grid was left blank for gas-phase measurements.
The oxide samples prepared on the tungsten grid are secured
inside the infrared cell by Teflon-coated sample holder jaws.
The sample cell is connected to a chamber through a Teflon
tube that contains ports for gas introduction and two absolute
pressure transducers (range 0.001-10.000 and 0.1-1000.0
Torr). The total volume of the infrared cell is 1214 ( 4 mL
(infrared cube 197 ( 2 mL and mixing chamber 1044 ( 1 mL).
The vacuum chamber consists of a two-stage pumping system,
a turbomolecular/mechanical pump for pumping to 10-7 Torr,
and a mechanical pump for rough pumping to 10-3 Torr. The
infrared cell is mounted on a linear translator inside the FT-IR
spectrometer. The translator allows both halves of the grid, the
blank side for gas-phase measurements and the oxide-coated
side for surface measurements, to be probed by simply moving
the infrared cell through the IR beam path. After the aluminum
oxide surface was exposed to relatively high pressures of nitric
acid (1.78 ( 0.01 Torr) for 30 min to saturate the surface with
nitrate, the FTIR cell was evacuated overnight to remove all
weakly adsorbed products.
output was measured using a spectroradiometer (model RPS900-
R, International Light Technologies). According to these
measurements, adsorbed nitrate receives broadband light ap-
proximately equal to one solar constant. After irradiation,
infrared spectra were recorded with a single-beam Mattson RS-
10000 spectrometer equipped with a narrow band MCT detector.
Typically, 250 scans were collected with an instrument resolu-
tion of 4 cm-1 in the spectral range extending from 900 to 4000
cm-1. We obtained absorbance spectra for gas and adsorbed
species by referencing single-beam spectra of the blank grid
and the oxide-coated grid to single beam spectra collected prior
to the nitric acid exposure. To investigate surface-mediated post
irradiation processes, the light source was turned off and IR
spectra (25 scans, 4 cm-1 resolution) were immediately collected
of the gas phase every 3 min for 2 h.
Ex Situ X-ray Photoelectron Spectroscopy. The irradiated
aluminum oxide sample was removed from the tungsten grid
and analyzed using a custom-designed Kratos Axis Ultra X-ray
photoelectron spectrometer with a monochromatic Al KR X-ray
source. The sample was pressed onto an indium foil on a copper
stub and introduced into the XPS analysis chamber, which had
a pressure that was maintained in the 10-9 Torr range during
analysis. Wide energy range survey scans are acquired using
the following parameters: energy range from 1200 to -5 eV,
pass energy of 160 eV, step size of 1 eV, dwell time of 200
ms, and X-ray spot size of 700 × 300 mm2. High-resolution
spectra are acquired using the following parameters: energy
range of 50 to 20 eV depending on the peak examined, pass
energy of 20 eV, step size of 0.1 eV, and dwell time of 1000
ms. The data collected are analyzed using CasaXPS data
processing software. The instrument and data reduction are
described in detail in ref 21.
Following the introduction of water vapor, molecular O2 (100
( 2 Torr), or both, the valve connecting the FTIR cell to the
mixing chamber was closed, letting the gas-phase products
accumulate inside the cell as the surface was irradiated. In these
studies, a 500 W broadband Hg arc lamp (Oriel, model no.
66033), followed by a water filter to remove IR radiation and
a filter to remove shorter wavelengths (λ < 300 nm, Oriel filter
no. 59425), was used to irradiate the surface. The intensity of
the broadband light source, reaching to the sample inside the
custom-made IR cell, was measured using a solar cell (model
EI-100, Optical Energy Technologies). Irradiance of the lamp
Results and Discussion
Fourier Transform Infrared Spectroscopy of Adsorbed
Nitrate before and after Irradiation as f(%RH). The trans-
mission FTIR spectra of γ-Al2O3 exposed to nitric acid vapor
followed by evacuation are shown in Figure 1a at four different

7820 J. Phys. Chem. A, Vol. 113, No. 27, 2009
Rubasinghege and Grassian
TABLE 1: Percent Nitrate on the Surface Following Broadband Irradiation (λ > 300 nm) under Different Conditionsa,b
irradiation time/min
%RH < 1
without O2
20 ( 2
45 ( 2
without O2
80 ( 2
with O2
with O2
with O2
with O2
30
120
420
95 ( 1
85 ( 1
82 ( 1
83 ( 3 (84 ( 1)c
61 ( 3 (65 ( 2)c
41 ( 8 (44 ( 3)c
89 ( 1
78 ( 3
74 ( 1
90 ( 1
80 ( 2
77 ( 2
72 ( 2 (74 ( 1)c
60 ( 2 (62 ( 1)c
40 ( 3 (43 ( 2)c
93 ( 1
82 ( 1
79 ( 1
a At t ) 0 min, there is 100% nitrate on the surface. b Reported nitrate loss is the mean of triplicate measurements, and error represents (1σ.
c In select experiments, nitrogen (N2) was used as a buffer gas at a pressure of 100 Torr, and these data are reported in parentheses. See the
text for further details.
relative humidities, <1, 20, 45, and 80%. Under conditions of
lowest RH (%RH < 1), adsorbed nitrate binds to the oxide
surface in different modes of coordination as monodentate,
bidentate, and bridging, as previously discussed in detail, on
the basis of the assignment of the peaks in the nitrate ion ν3
spectral region between 1200 and 1650 cm-1.4,7 In the presence
of water vapor at different relative humidities (20, 45, and 80%),
co-adsorbed water solvates the nitrate ion, forming inner and
outer sphere complexes that are represented by peaks in the
region between 1300 and 1450 cm-1 in Figure 1a.7 Changes in
the infrared spectra as a function of RH and the assignment of
these peaks have been discussed extensively, and the details
can be found in refs 4 and 7.
Following the adsorption of nitric acid and the introduction
of water vapor adjusted to the different relative humidities, the
Figure 2. Photo-induced surface nitrate loss in the presence of
sample was irradiated with the broadband light (λ > 300 nm).
molecular oxygen under different relative humidity conditions (RH <
Upon irradiation, the FTIR spectra of the nitrated surface at
each RH were recorded. It can be seen in Figure 1b that the
spectrum changes after 420 min of irradiation. Most notably,
1, 20, 45, and 80%). θo-adsorbed nitrate concentration on saturated
oxide surface at t ) 0. θ-nitrate concentration on oxide surface at
irradiation time, t. The presented nitrate loss is the mean of triplicate
there is a decrease in the intensity of the spectral bands
associated with adsorbed nitrate. An analysis of the changes in
the intensities of these bands, including an analysis of the
difference spectra, shows that there is a loss of intensity at low
RH near 1559 cm-1, assigned to the monodentate adsorbed
nitrate, and the bands at 1588 and 1307 cm-1, assigned to
bridging nitrate. That is to say that the monodentate species
shows the greatest loss upon irradiation, followed by bridging
nitrate absorptions. This reveals that there is a greater propensity
for certain oxide-coordinated adsorbed nitrates to undergo
photolysis. At higher RH, the solvated nitrate peaks, observed
between 1300 and 1450 cm-1, all decrease upon irradiation.
The percent loss of surface nitrate as a function of irradiation
time is given in Table 1. The percent loss is determined using
normalized integrated absorbance of the entire adsorbed nitrate
ν3 region extending from ca. 1220 to 1700 cm-1 for %RH < 1
and from 1200 to 1550 cm-1 for solvated nitrate at higher RH.
The effect of molecular oxygen on nitrate photolysis was studied
under two different RH conditions: %RH < 1 and %RH 45 (
2. In both the cases, the surface nitrate loss decreased signifi-
cantly in the presence of molecular oxygen. As shown in Table
1, under dry conditions, the total nitrate on the surface decreases
from 100% (at time t ) 0) to 82 ( 1 and 41 ( 8% in the
presence and absence of molecular oxygen, respectively, fol-
lowing 420 min of irradiation. Similar changes are observed
under humid conditions, 77 ( 2 and 40 ( 3%, upon irradiation
in the presence and absence of molecular oxygen, respectively.
Control experiments in the presence of molecular nitrogen reveal
that changes in the amount of surface nitrate loss are a result of
interaction of the surface with molecular oxygen and not just
due to the presence of a buffer gas. In a number of experimental
and theoretical studies, it has been reported that defect sites
including the presence of neutral oxygen vacancies and charged
oxygen vacancies are present on aluminum oxide.22 These
measurements, and the error bars represent the standard deviation.
O(vacancy) sites may play a role, albeit unknown, in the primary
photochemical process of adsorbed nitrate or the chemistry of
photoproducts. If these sites are primary binding sites for the
adsorption of molecular oxygen, then oxygen will have a large
impact on the photochemistry of adsorbed nitrate, as is evident
in these studies.
As seen in Table 1, the effect of RH on adsorbed nitrate
photolysis is more significant in the presence of molecular
oxygen. In the oxidative environment provided by the presence
of molecular oxygen, a higher nitrate loss was recorded under
humid conditions compared with that of dry (%RH < 1). To
understand the effect of RH on nitrate photolysis further, more
experiments were carried out under additional conditions of RH
in the presence of molecular oxygen. The higher loss of surface
nitrate under humid conditions can be attributed to the formation
of inner and outer sphere complexes of nitrate ion, which have
less interaction with the surface and less molecular distortion
from the planar D3h symmetry of the free ion compared with
the oxide coordinated nitrate structures present under dry
conditions.7 Furthermore, in the presence of water vapor, there
is a possibility of forming hydroxyl radicals as a result of nitrate
photolysis.14,23 These hydroxyl radicals can initiate new pho-
tochemical pathways leading to a higher nitrate loss from the
oxide surface.24 H-bonding between solvated nitrate and coad-
sorbed water affects the energetics of the electronic n-to-π*
transition near 300 nm with a red shift resulting in a higher
overlap of the absorption profile of nitrate with the broadband
irradiation source.13,25
According to Figure 2, during several hours of photolysis of
adsorbed nitrate, the change in the rate of nitrate disappearance
is most significant in the first 90 min. An attempt to fit the time
dependence of nitrate photolysis to first-order kinetics to gain

Photochemistry of Adsorbed Nitrate
J. Phys. Chem. A, Vol. 113, No. 27, 2009 7821
Figure 3. FTIR spectra of gas-phase products formed during photolysis of adsorbed nitrate under dry conditions (%RH < 1) at 298 K (a) in the
absence of molecular oxygen and (b) in the presence of molecular oxygen (100 ( 5 Torr). Each spectrum was referenced to the gas-phase spectrum
prior to exposure to nitric acid.
an estimate of an apparent rate constant is shown in the inset
of Figure 2. According to this analysis, the first-order photolysis
rate constants are (3.2 ( 0.2) × 10-5, (5.6 ( 0.3) × 10-5, (4.9
( 0.2) × 10-5, and (4.2 ( 0.2) × 10-5 s-1 for RH < 1, 20, 45,
and 80%, respectively. Therefore, the initial rate of adsorbed
nitrate photolysis is approximately 1.5 to 2 times faster in the
presence of co-adsorbed water.
consider what is known about nitrate photochemistry in solution
phase and snowpack. It is well known that irradiation of the
nitrate ion with light at 305 nm under acidic conditions yields
nitrogen dioxide and nitrite. Further photolysis of the nitrite ion
at 365 nm produces NO,14,16-19 according to the following
reactions
Along with the photochemical loss of nitrate, there is the
growth of product bands as well. As seen in the spectra shown
in Figure 1b, there is a peak near 1702 cm-1 that grows in
intensity, under all conditions of RH. The presence of the band
at 1702 cm-1 on the surface suggests the formation of a new
surface product. The assignment of this band as well as the
assignment of several nitrogen-containing gas-phase products
are discussed in detail below.
NO3- + H+ 305 nm8 NO2 + OH
(1)
(2)
NO-3 305 nm8 NO2- + O(3P)
and
Analysis of Gas-Phase and Surface-Bound Products. As
shown in Figure 3, gas-phase products appear as a function of
irradiation of adsorbed nitrate. FTIR spectra of these gas-phase
products recorded as a function of irradiation time under dry
conditions (%RH < 1) in both the presence and absence of
molecular oxygen are shown. It is relatively easy to identify
the major gas-phase products to be N2O, NO, and NO2 with
NO2- + H+ 365 nm8 NO + OH
(3)
The current results for the surface photochemistry of adsorbed
nitrate can begin to be explained in the framework of nitrate
ion photochemistry in acidic aqueous media, as shown by
reactions 1-3. Figure 4 shows that during the first 30 min of
irradiation, adsorbed nitrate yields NO2 as the major gas-phase
product in the presence of co-adsorbed water. However, for
longer irradiation times, that is, after 120 and 420 min, the major
gas-phase product is seen to be NO, followed by NO2 and N2O.
This change in gas-phase product distribution may partially be
associated with gas-phase NO2 photochemistry to yield NO. In
a reaction cell with only gas-phase NO2 present, photolysis is
measured and determined to be 200-fold faster compared with
adsorbed nitrate photolysis.
In addition, N2O production is higher under the highest RH
conditions. These variations in the gas-phase product distribution
can be seen more clearly in the plots shown in Figure 5. These
plots, showing the relative gas-phase concentrations of NO2,
N2O, and NO at short (30 min), medium (120 min), and longer
(420 min) irradiation times, were obtained by normalizing the
gas-phase concentrations to the largest gas-phase product species
in a given RH, which is NO at long times in all cases.
As can be seen in the data shown in Figures 4 and 5, the
gas-phase NO2 concentration increases to %RH 20 and slightly
decreases after %RH 45. Under higher RH conditions, that is,
characteristic absorptions at 2224, 1872, and 1614 cm-1
,
respectively.20 A comparison of the spectra shown in Figure
3a, b reveals a significant change in the gas-phase product
distribution, in particular the relative N2O and NO2 concentra-
tions under these two different environmental conditions. This
large difference was noted and previously discussed in our
earlier study.20 Therefore, in this work we focus more on gas
product distribution under different RH conditions in an
oxidative environment, that is, in the presence of molecular
oxygen, because this is more atmospherically relevant conditions.
Gas-phase concentrations obtained from FTIR analysis,
followed by broadband (λ > 300 nm) irradiation of adsorbed
nitrate in the presence of molecular oxygen under the different
RH conditions (RH < 1, 20, 45, and 80%) are shown in Figure
4. The results presented in Figure 4 are the average value of
triplicate measurements, and the reported error represents one
standard deviation. We determined gas-phase concentrations of
individual species by converting integrated absorbances to
concentrations using calibration factors for the individual gas-
phase species. These data show that gas-phase NO2, NO, and
N2O are produced at different rates and in different yields.
Before further discussing the photochemistry of adsorbed
nitrate and the photoproducts that form, it is first instructive to

7822 J. Phys. Chem. A, Vol. 113, No. 27, 2009
Rubasinghege and Grassian
Figure 4. Gas-phase concentrations obtained from FTIR spectra following irradiation of adsorbed nitrate in the presence of molecular oxygen at
298 K under different relative humidity (%RH) conditions: (a) <1, (b) 20 ( 2, (c) 45 ( 2, and (d) 80 ( 2.
Figure 5. Relative gas-phase concentrations obtained from FTIR spectra following irradiation of adsorbed nitrate in the presence of molecular
oxygen at 298 K under different relative humidity (%RH) conditions: (a) <1, (b) 20 ( 2, (c) 45 ( 2, and (d) 80 ( 2 at short (30 min), medium
(120 min), and longer (420 min) irradiation times. The gas-phase concentrations have been normalized to the largest gas-phase product species after
420 min of irradiation.
%RH 45 and 80, there is a possibility to react NO2 with co-
adsorbed water to yield solvated nitrate on the surface.26,27 The
readsorption of gas-phase products is discussed in detail in the
discussion of post-photolysis processes. Photolysis of NO2- to
NO is not predominant under humid conditions during the first
30 min of irradiation. Formation of [NO2- ·H2O] in the presence

Photochemistry of Adsorbed Nitrate
J. Phys. Chem. A, Vol. 113, No. 27, 2009 7823
of co-adsorbed water and lower acidity may be the possible
reasons for the above observation. In addition, protons (H+)
generated from adsorbed water are utilized in a competitive
reaction during the formation of N2O under these conditions,
as discussed below. Furthermore, as irradiation proceeds, the
accumulation of nitrite may favor the production of NO.
Although attempts to detect surface-adsorbed nitrite were
unsuccessful, it potentially may be present at very low level
steady-state coverages under these conditions.
As we previously reported, the formation of nitrous oxide
(N2O) is observed during the photolysis of adsorbed nitrate. The
relative N2O concentration is highly significant in the absence
of molecular oxygen and also contributes to a major fraction in
the presence of oxygen under humid conditions at the end of
7 h of photolysis. As we have previously discussed, surface-
catalyzed reactions of NO can lead to the formation of N2O,
which is easily quenched in the presence of molecular oxygen.28
Malecki and Malecka, in their studies on thermal decomposition
of metal nitrates, have proposed two other possible reaction
mechanisms for the formation of N2O in the presence of NO
and NO2.29
Figure 6. Schematic of the reaction mechanism of heterogeneous
photolysis of adsorbed nitrate. HNO3(g) (left side of the diagram)
adsorbs from the gas phase onto the surface to yield nitrate ion
(NO3-(a)) and protons (H+) in the presence and absence of coadsorbed
water. Broadband irradiation results in the production of gas-phase and
surface-adsorbed products, as shown by the different species. The
observed gas-phase and surface-bound products in the transmittance
FTIR spectroscopy are colored green, whereas gas-phase and surface-
bound proposed intermediates not detected by FTIR spectroscopy
Mechanism 1 NO2 + NO-3 f N2O + 7/4 O2 + 1/2 O2-
(4)
-
including gas-phase N2O2 and HONO and adsorbed NO2 are shown
in white.
Mechanism 2 2NO f N2O2
N2O2 + NO f N2O + NO2
(5)
(6)
RH conditions at which the N2O production is highest. The
formation of HONO and its photochemical products are further
discussed below.
In the presence of oxygen, N2O2 can be oxidized to yield
NO2
Because NO is the major gas-phase product, there is a
possibility of readsorbing NO on the aluminum oxide surface.
In previous studies, vibrational modes of adsorbed NO have
been reported and assigned to bands at ∼1700-1720 cm -1 on
different surfaces.37-40 The adsorption band near 1702 cm-1 in
the adsorbed phase spectra that grows during irradiation is most
likely due to NO adsorbed to oxide surfaces. Furthermore, the
X-ray photoelectron spectrum (XPS) in the N(1s) region of the
irradiated nitrated aluminum oxide surface revealed a peak at
401 eV, which was not observed in the XPS of samples exposed
to nitric acid and not irradiated. The peak at 401 eV is consistent
with the presence of adsorbed NO on the surface.41 For both
samples that were just exposed to nitric acid and those exposed
to nitric acid and irradiated, a peak at 407 eV was present in
the N(1s) region. This higher energy peak is consistent with
adsorbed nitrate on the surface. Higher production of NO2 in
the presence of O2 can lead to NO2 adsorption on the surface
to yield adsorbed nitrates, thus contributing as another source
to the overall nitrate on the surface.
In recent studies, evidence of photochemical generation of
nitrous acid, HONO, from the surface reaction of hydroxyl
radicals with nitrogen dioxide in the presence of water vapor is
available.42-46 Furthermore, it is reported that HONO flux is
wavelength dependent, and its release is strongest when λ <
320 nm, where the absorption of aqueous nitrate is highest.47
Again, gas-phase HONO can be photochemically converted back
to NO and OH.42 In these studies HONO was not observed in
the gas phase, although its photolysis product NO is readily
observed.
N2O2 + O2 f 2NO2
(7)
In addition, Maric et al. also have shown a similar mechanism
for the formation of N2O via a heterogeneous dark conversion
of NO2, N2O5, or both.30 The above two pathways provide
possible mechanisms for the production of N2O and changes in
the relative gas-phase concentrations of N2O and NO2, respec-
tively, in the presence of molecular oxygen under dry conditions.
At higher relative humidities, particularly at %RH 45 and 80,
a significant amount of gas-phase N2O was observed, even in
an oxidizing environment. In previous studies on heterogeneous
hydrolysis of NO2, it has been shown that N2O is produced under
humid conditions and longer reaction times.27 Recently, Wiesen
et al. and Kleffmann et al. have described a possible reaction
mechanism for the formation of N2O via HONO in their studies
on the hydrolysis of NO2 on an acidic surface.31-33 The overall
reaction of the proposed mechanism has been described as
8NO2 + 3H2O [H + ]8 N2O + 6HNO3
(8)
In this reaction pathway, the hyponitrous acid (HONdNOH)
intermediate is proposed to decompose to N2O over a wide pH
range, including acidic conditions.34-36 These mechanisms and
the pathways discussed for NO2 surface reactions under humid
conditions suggest that the higher relative gas-phase concentra-
tion of N2O is due to a secondary reaction of NO2 on the acidic
oxide surface. This can also be one of the potential reasons for
the observed lower gas-phase NO2 concentrations under extreme
The schematic of the proposed reaction mechanism of
heterogeneous photolysis of adsorbed nitrate is shown in Figure
6. Heterogeneous uptake of nitric acid, HNO3(g), on the left
side of the diagram yields nitrate ion (NO3-(a)) and protons

7824 J. Phys. Chem. A, Vol. 113, No. 27, 2009
Rubasinghege and Grassian
(H+) on the surface in the presence and absence of coadsorbed
water. Upon broadband irradiation, NO and NO2 are produced
as major gas-phase products. Photochemistry of gas-phase NO2
yields NO, which makes NO the dominant product. In the
absence of O2, NO acts as the precursor for N2O via a surface-
mediated reaction, whereas the reaction mechanism is poisoned
in the presence of molecular oxygen, resulting in more NO2
and less N2O. In the presence of coadsorbed water, NO2
becomes the precursor for gas-phase N2O through heterogeneous
hydrolysis of NO2 on the acidic surface. The proposed mech-
anism involves gas-phase species of N2O2 and HONO, whereas
NO2-(a) is an intermediate of gas-phase NO formation via
photolysis. In addition, the gas-phase species are readsorbed
onto the surface as nitrate and NO(a).
Dark Reaction: Gas-Phase Product Distribution Changes
Postphotolysis Involving a Surface-Catalyzed Reaction. Rela-
tive gas-phase distributions presented in the last section for %RH
< 1 and %RH 45 differ from that reported in our previous work.
In the earlier work, following photolysis, the lamp was turned
off and the FTIR sample compartment was purged for several
minutes prior to recording a spectrum so as to remove CO2 and
H2O from the FTIR compartment with dry air. In the current
study, the apparatus was modified so that the lamp was
incorporated into the dry air purge box so that extra purge time
post-photolysis was unnecessary in these current experiments.
During the time it took to purge the compartment, there were
changes in the gas-phase product distribution, unbeknownst in
the earlier studies. This change in the gas-phase distribution
observed here led us to investigate this postphotolysis dark
reaction further.
FTIR spectra collected every 3 min of the gas phase
immediately after the broadband light source was turned off
are shown in Figure 7a. As seen in the spectra, the N2O
concentration remains constant with time, whereas the NO
concentration decreases concomitant with an increase in the
NO2 concentration. As shown in Figure 5b, at RH < 1%, the
NO2 concentration reaches a maximum after around 100 min,
and NO decays during this period. The same effect was observed
under all humidity conditions investigated. However, at higher
RH, NO2 reached a maximum in a shorter time period. The
basic chemical reaction most likely responsible for this gas-
phase conversion is
Figure 7. Changes in the gas-phase product distribution during post-
irradiation of adsorbed nitrate. (a) FTIR spectra of gas-phase products
collected under %RH < 1 following irradiation for 420 min. (b) Change
of gas-phase product (NO and NO2) concentrations as a function of
time under %RH < 1 following irradiation. (c) Relative gas-phase
concentrations after 120 min of post-photolysis under different RH
conditions.
2NO + O2 f 2NO2
(8a)
The oxidation of NO to NO2 is an important intermediate
step involved in NOx abatement that is being developed for
diesel engines.48 It is also the first step that occurs in the cyclic
NOx storage and recycle process.49
Relative gas-phase distributions after 120 min of post-
photolysis under different experimental conditions are shown
in Figure 5c. In contrast with the distributions given in Figure
5, NO2 is the dominant product in the dark under the most
atmospherically relevant conditions, that is, under humid
conditions and in the presence of molecular oxygen. The relative
gas-phase concentration of N2O appears to be increasing under
humid conditions.
photolysis of solvated nitrate is higher than that of oxide-
coordinated nitrate under dry conditions. This could be a
potentially important, yet previously unknown, surface-mediated
photochemical pathway in the NOy, cycle. Here we reported
the production of N2O, an important greenhouse gas, with higher
concentrations under humid conditions in the presence of
molecular oxygen. We propose that this reaction involves a
heterogeneous conversion of gas-phase NO and NO2. Higher
relative gas-phase concentrations of N2O under humid conditions
can be attributed to a secondary reaction of NO2 on acidic oxide
surfaces. The present work also focused on the chemistry of
surface-mediated gas-phase product distribution changes. In the
dark, gas-phase NO is oxidized to NO2 in the presence of O2,
whereas the N2O concentration remains constant. The rate of
this gas-phase conversion increases as a function of RH because
of increased readsorption of NO2 back to the surface under
humid conditions. The higher oxidation rates of NO by
molecular oxygen suggest an involvement of the oxide surface
Conclusions and Atmospheric Implications
The photochemical conversion of nitric acid to gas-phase
nitrous oxide, nitric oxide, and nitrogen dioxide through an
adsorbed nitrate intermediate under different atmospherically
relevant conditions has been shown. Under humid conditions,

Photochemistry of Adsorbed Nitrate
J. Phys. Chem. A, Vol. 113, No. 27, 2009 7825
(14) Goldstein, S.; Rabani, J. J. Am. Chem. Soc. 2007, 129, 10597.
(15) Young, M. A. Environmental Photochemistry in Surface Waters.
In Water Encyclopedia; John Wiley & Sons: New York, 2005.
(16) Boxe, C. S.; Colussi, A. J.; Hoffmann, M. R.; Perez, I. M.; Murphy,
J. G.; Cohen, R. C. J. Phys. Chem. A 2006, 110, 3578.
(17) Chu, L.; Anastasio, C. J. Phys. Chem. A 2003, 107, 9594.
(18) Mack, J.; Bolton, J. R. J. Photochem. Photobiol., A 1999, 128, 1.
(19) Zellner, R.; Exner, M.; Herrmann, H. J. Atmos. Chem. 1990, 10,
411.
(20) Schuttlefield, J.; Rubasinghege, G.; El-Maazawi, M.; Bone, J.;
Grassian, V. H. J. Am. Chem. Soc. 2008, 130, 12210.
(21) Balrusaitis, J.; Usher, C. R.; Grassian, V. H. Phys. Chem. Chem.
Phys. 2007, 9, 3011.
(22) Evarestor, R. A. Quantum Chemistry of Solids: The LCAO First
Principles Treatment of Crystals; Springer: New York, 2007.
(23) Vione, D.; Maurino, V.; Minero, C.; Pelizzetti, E.; Harrison,
M. A. J.; Olariu, R. I.; Arsene, C. Chem. Soc. ReV. 2006, 35, 441.
(24) Gonza´lez, M. C.; Roma´n, E. S. Environmental Photochemistry in
Heterogeneous Media. In EnVironmental Photochemistry, Part II; Springer:
Berlin, 2005; Vol. 2, p 49.
(25) Miller, T. M.; Grassian, V. H. Geophys. Res. Lett. 1998, 25, 3835.
(26) Cheung, J. L.; Li, Y. Q.; Boniface, J.; Shi, Q.; Davidovits, P.;
Worsnop, D. R.; Jayne, J. T.; Kolb, C. E. J. Phys. Chem. A 2000, 104,
2655.
(27) Finlayson-Pitts, B. J.; Wingen, L. M.; Sumner, A. L.; Syomin, D.;
Ramazan, K. A. Phys. Chem. Chem. Phys. 2003, 5, 223.
(28) Trogler, W. C. Coord. Chem. ReV. 1999, 187, 303.
(29) Malecki, A.; Malecka, B. Thermochim. Acta 2006, 446, 113.
(30) Maric, D.; Burrows, J. P.; Moortgat, G. K. J. Atmos. Chem. 1992,
15, 157.
(31) Wiesen, P.; Kleffmann, J.; Kurtenbach, R.; Becker, K. H. Faraday
Discuss. 1995, 100, 121.
(32) Kleffmann, J.; Becker, K. H.; Wiesen, P. Atmos. EnViron. 1998,
32, 2721.
in the reaction mechanism. Therefore, we propose a surface-
mediated dark reaction for the above gas-phase conversion.
These results have implications for atmospheric processes
involving nitric acid, NOx, and mineral dust aerosol. Further-
more, the results provide some evidence of photochemical
reactions of atmospheric significance taking place on the surface
of particulate matter. This study also reveals that in the
troposphere, NOx level can be affected by these continued
reactions of adsorbed nitrate initiated by sunlight. Moreover,
the heterogeneous photochemical conversion of nitric acid to
NOx could contribute to renoxification of the atmosphere. In
atmospheric field studies, the interpretation of observed trends
in terms of reaction mechanisms is difficult because of the
involvement of a complex blend of substrates distributed in
several phases and the lack of knowledge in composition of
particulate matter such as mineral dust and sea salt and their
behavior under different atmospheric conditions. Therefore,
detailed studies, such as this one of potential reactions involving
heterogeneous photochemistry of associated secondary species
such as nitrate on aerosols, may be useful in better explaining
field results of atmospheric chemistry and may be implemented
in atmospheric chemistry models that can be used to explain
and predict concentrations of key gas-phase constituents under
a variety of environmental conditions.
Acknowledgment. This material is based on the work
supported by the National Science Foundation under grant
CHE0503854. Any opinions, findings, and conclusions or
recommendations expressed in this material are those of authors
and do not necessarily reflect the views of the National Science
Foundation.
(33) Kleffmann, J.; Becker, K. H.; Wiesen, P. J. Chem. Soc., Faraday
Trans. 1998, 94, 3289.
(34) Hughes, M. N.; Stedman, G. J. Chem. Soc. 1964, 163.
(35) Akhtar, M. J.; Balschi, J. A.; Bonner, F. T. Inorg. Chem. 1982, 21,
2216.
(36) Loechler, E. L.; Schneider, A. M.; Schwartz, D. B.; Hollocher, T. C.
J. Am. Chem. Soc. 1987, 109, 3076.
(37) Berlier, G.; Spoto, G.; Ricchiardi, G.; Bordiga, S.; Lamberti, C.;
Zecchina, A. J. Mol. Catal A: Chem. 2002, 182, 359.
(38) Kazusaka, A.; Suzuki, H.; Toyoshima, I. Chem. Commun. 1983,
150.
(39) Hu, Y.; Griffiths, K. Surf. Sci. 2008, 602, 2949.
(40) Hu, Y. H.; Griffiths, K. Appl. Surf. Sci. 2008, 254, 1666.
(41) Overbury, S. H.; Mullins, D. R.; Huntley, D. R.; Kundakovic, L.
J. Catal. 1999, 186, 296.
(42) Ramazan, K. A.; Syomin, D.; Finlayson-Pitts, B. J. Phys. Chem.
Chem. Phys. 2004, 6, 3836.
(43) Ndour, M.; D’Anna, B.; George, C.; Ka, O.; Balkanski, Y.;
Kleffmann, J.; Stemmler, K.; Ammann, M. Geophys. Res. Lett. 2008, 35,
L05812.
(44) Gustafsson, R. J.; Orlov, A.; Griffiths, P. T.; Cox, R. A.; Lambert,
R. M. Chem. Commun. 2006, 3936.
(45) Stemmler, K.; Ammann, M.; Donders, C.; Kleffmann, J.; George,
C. Nature 2006, 440, 195.
(46) George, C.; Strekowski, R. S.; Kleffmann, J.; Stemmler, K.;
Ammann, M. Faraday Discuss. 2005, 130, 195.
(47) Bartels-Rausch, T.; Donaldson, D. J. Atmos. Chem. Phys. Discuss.
2006, 6, 10713.
References and Notes
(1) Usher, C. R.; Michel, A. E.; Grassian, V. H. Chem. ReV. 2003,
103, 4883.
(2) Laskin, A.; Iedema, M. J.; Cowin, J. P. EnViron. Sci. Technol. 2002,
36, 4948.
(3) Kato, N.; Akimoto, H. Atmos. EnViron., Part A 1993, 27, 1163.
(4) Goodman, A. L.; Bernard, E. T.; Grassian, V. H. J. Phys. Chem. A
2001, 105, 6443.
(5) Seinfeld, J. H., Pandis, S. N. Atmospheric Chemistry and Physics
From Air Pollution to Climatic Change, 2nd ed.; John Wiley and Sons:
New York, 2006.
(6) Sillman, S.; Logan, J. A.; Wofsy, S. C. J. Geophys. Res., [Atmos.]
1990, 95, 1837.
(7) Baltrusaitis, J.; Schuttlefield, J.; Jensen, J. H.; Grassian, V. H. Phys.
Chem. Chem. Phys. 2007, 9, 4970.
(8) Gomes, L.; Gillette, D. A. Atmos. EnViron., Part A 1993, 27, 2539.
(9) Warneck, P. Chemistry of the Natural Atmosphere; Academic Press:
San Diego, CA, 1988; Vol. 41.
(10) Angelini, M. M.; Garrard, R. J.; Rosen, S. J.; Hinrichs, R. Z. J.
Phys. Chem. A 2007, 111, 3326.
(11) Mashburn, C. D.; Frinak, E. K.; Tolbert, M. A. J. Geophys. Res.,
[Atmos.] 2006, 111, D15213.
(12) Hatch, C. D.; Gough, R. V.; Toon, O. B.; Tolbert, M. A. J. Phys.
Chem. B 2008, 112, 612.
(13) Cwiertny, D. M.; Young, M. A.; Grassian, V. H. Annu. ReV. Phys.
Chem. 2008, 59, 27.
(48) Xue, E.; Seshan, K.; Vanommen, J. G.; Ross, J. R. H. Appl. Catal.,
B 1993, 2, 183.
(49) Castoldi, L.; Matarrese, R.; Lietti, L.; Forzatti, P. Appl. Catal., B
2006, 64, 25.
JP902252S