New experimental and theoretical approach to the heterogeneous hydrolysis of NO2: Key role of molecular nitric acid and its complexes

Article abstract of DOI:10.1021/jp056426n

Although heterogeneous chemistry on surfaces in the troposphere is known to be important, there are currently only a few techniques available for studying the nature of surface-adsorbed species as well as their chemistry and photochemistry under atmospheric conditions of 1 atm pressure and in the presence of water vapor. We report here a new laboratory approach using a combination of long path Fourier transform infrared spectroscopy (FTIR) and attenuated total reflectance (ATR) FTIR that allows the simultaneous observation and measurement of gases and surface species. Theory is used to identify the surface-adsorbed intermediates and products, and to estimate their relative concentrations. At intermediate relative humidities typical of the tropospheric boundary layer, the nitric acid formed during NO₂ heterogeneous hydrolysis is shown to exist both as nitrate ions from the dissociation of nitric acid formed on the surface and as molecular nitric acid. In both cases, the ions and HNO₃ are complexed to water molecules. Upon pumping, water is selectively removed, shifting the NO₃^--HNO ₃(H₂O)_y, equilibria toward more dehydrated forms of HNO₃ and ultimately to nitric acid dimers. Irradiation of the nitric acid-water film using 300-400 nm radiation generates gaseous NO, while irradiation at 254 nm generates both NO and HONO, resulting in conversion of surface-adsorbed nitrogen oxides into photochemically active NO_x, These studies suggest that the assumption that deposition or formation of nitric acid provides a permanent removal mechanism from the atmosphere may not be correct. Furthermore, a potential role of surface-adsorbed nitric acid and other species formed during the heterogeneous hydrolysis of NO₂ in the oxidation of organics on surfaces, and in the generation of gas-phase HONO on local to global scales, should be considered.

Full text of DOI:10.1021/jp056426n

6886
J. Phys. Chem. A 2006, 110, 6886-6897
New Experimental and Theoretical Approach to the Heterogeneous Hydrolysis of NO2: Key
Role of Molecular Nitric Acid and Its Complexes†
K. A. Ramazan,‡,§ L. M. Wingen,‡ Y. Miller,| G. M. Chaban, R. B. Gerber,‡,|
S. S. Xantheas,# and B. J. Finlayson-Pitts*,‡
Department of Chemistry, UniVersity of California, IrVine, California 92697-2025
ReceiVed: NoVember 7, 2005; In Final Form: December 31, 2005
Although heterogeneous chemistry on surfaces in the troposphere is known to be important, there are currently
only a few techniques available for studying the nature of surface-adsorbed species as well as their chemistry
and photochemistry under atmospheric conditions of 1 atm pressure and in the presence of water vapor. We
report here a new laboratory approach using a combination of long path Fourier transform infrared spectroscopy
(FTIR) and attenuated total reflectance (ATR) FTIR that allows the simultaneous observation and measurement
of gases and surface species. Theory is used to identify the surface-adsorbed intermediates and products, and
to estimate their relative concentrations. At intermediate relative humidities typical of the tropospheric boundary
layer, the nitric acid formed during NO2 heterogeneous hydrolysis is shown to exist both as nitrate ions from
the dissociation of nitric acid formed on the surface and as molecular nitric acid. In both cases, the ions and
-
HNO3 are complexed to water molecules. Upon pumping, water is selectively removed, shifting the NO3 -
HNO3(H2O)y equilibria toward more dehydrated forms of HNO3 and ultimately to nitric acid dimers. Irradiation
of the nitric acid-water film using 300-400 nm radiation generates gaseous NO, while irradiation at 254 nm
generates both NO and HONO, resulting in conversion of surface-adsorbed nitrogen oxides into photochemically
active NOx. These studies suggest that the assumption that deposition or formation of nitric acid provides a
permanent removal mechanism from the atmosphere may not be correct. Furthermore, a potential role of
surface-adsorbed nitric acid and other species formed during the heterogeneous hydrolysis of NO2 in the
oxidation of organics on surfaces, and in the generation of gas-phase HONO on local to global scales, should
be considered.
Introduction
there is an organic layer on the surfaces of buildings and that
the composition of this layer changes with location as the air
masses move downwind. A particularly striking example of
heterogeneous tropospheric chemistry is the hydrolysis of NO2
on surfaces, which has been recognized for many decades to
generate gas-phase HONO:5-26
Heterogeneous chemistry is important in both the lower and
upper atmosphere.1 In the stratosphere, for example, reactions
occurring on polar stratospheric clouds were shown about two
decades ago to play a major role in the rapid destruction of
ozone at polar sunrise in the Antarctic. Current understanding
of heterogeneous reactions in the troposphere is more limited
than that for the upper atmosphere, due in part to the wide range
of surfaces available for this chemistry, the thousands of
compounds present in air, and the variability in such parameters
as water vapor concentration, solar intensity, and meteorological
conditions.
2NO2 + H2O surface8 HONO + HNO3
(1)
This reaction has been extensively studied because the HONO
generated is a major source of OH not only at dawn, but also
when averaged over an entire day.27-36
Despite these complexities, there is increasing evidence that
heterogeneous chemistry is important in the troposphere, and
in particular in the boundary layer where increased surface area
is available in the form of buildings, roads, soil, vegetation,
etc. For example, Diamond and co-workers2-4 have shown that
Much less attention has been paid to the nitric acid produced
in reaction 1. Because it was not observed in the gas phase,
which could be more easily monitored in the early studies, HNO3
was assumed to remain on the surface. This was supported, for
example, by measurements of nitrate ions in water rinses of
the chamber walls after reaction.14 The presence of molecular
(undissociated) HNO3 was subsequently observed experimen-
tally using high surface area silica to follow the adsorbed
products by infrared spectroscopy.37,38 If nitric acid remains on
the surface, reaction 1 will contribute to the permanent removal
of oxides of nitrogen from the atmosphere. Nitric acid generated
in the gas phase is removed by wet and dry deposition, either
directly or after conversion to particulate ammonium nitrate.
It is commonly assumed that nitric acid deposited on surfaces
containing water will dissociate to H3O+ and NO3-. However,
recent laboratory studies suggest that, under some conditions,
† Part of the special issue “David M. Golden Festschrift”.
* To whom correspondence should be addressed. Phone: (949) 824-
7670. Fax: (949) 824-3168. E-mail: bjfinlay@uci.edu.
‡ Department of Chemistry, University of California Irvine, Irvine, CA
92697-2025.
§ Current address: KLA-Tencor Corporation, One Technology Drive,
Milpitas, CA 95035.
| Department of Physical Chemistry and the Friz Haber Research Center,
The Hebrew University, Jerusalem, Israel 91904.
NASA Ames Research Center, Mail Stop T27B-1, Moffet Field, CA
94035.
# Chemical Sciences Division, Pacific Northwest National Laboratory,
906 Battelle Boulevard, P. O. Box 999, MS K1-83, Richland, WA 99352.
10.1021/jp056426n CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/23/2006

Heterogeneous Hydrolysis of NO2
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6887
Figure 1. Schematic of the long path-attenuated total reflectance (LP-ATR) apparatus.
the nitric acid adsorbed on the surface is not entirely in the
form of nitrate ions, but rather is also in the form of nitric acid-
water complexes.26,39,40 Under some (but not necessarily all)41
conditions, these complexes can react with NO to generate NO2,
hence recycling the surface-adsorbed nitrate back into photo-
chemically active forms of NOx.39,42,43 In addition, photolysis
of nitric acid adsorbed on surfaces in the presence of water vapor
has been reported to generate gas-phase NO2, smaller amounts
of NO, and HONO.44-47 This has been hypothesized to be a
potentially significant source of daytime HONO. For example,
a large midday source of HONO (∼500 ppt h-1) was reported
in a recent field study.48 Finally, there is an “unknown” radical
source in large atmospheric smog chambers, which is thought
to be at least in part HONO,49-54 and which may be due to
photolysis of surface-adsorbed species such as nitric acid.
The majority of laboratory studies directed at elucidating the
kinetics, mechanisms, and products of reaction 1 have focused
on monitoring either the gas-phase products (e.g., HONO
formation) or, in fewer studies, the adsorbed products. These
were typically carried out using high surface area materials to
increase the sensitivity for surface species.
We describe here a new apparatus designed to follow
simultaneously both the gas-phase and adsorbed reaction
products in water films on substrates characteristic of surfaces
found in many laboratory systems as well as in the tropospheric
boundary layer. This is achieved by coupling in a unique
apparatus long path Fourier transform infrared spectroscopy (LP-
FTIR) for gas-phase components with an attenuated total
reflectance FTIR (ATR-FTIR) probe to follow the chemistry
occurring in the surface thin film. Sayer and Horn55 have
described an apparatus that also combines ATR with gas-phase
infrared spectroscopic detection, but there are significant dif-
ferences that are described in more detail below. Our new
apparatus is applied to the heterogeneous hydrolysis of NO2,
reaction 1, and theory is used to understand the nature of the
surface-adsorbed products. Finally, photolysis of the surface
products formed in reaction 1 in two different wavelength ranges
is reported. The atmospheric implications are discussed.
We are delighted to contribute this paper in honor of David
M. Golden and his 70th birthday, especially since Dave has
been a pioneer in laboratory studies of heterogeneous processes
in the atmosphere. His development of the Knudsen cell as a
tool for studying the uptake of gases on solids and low-volatility
liquids set the stage for a large number of studies that followed
using this technique. A significant fraction of the heterogeneous
atmospheric kinetics known today is based on such Knudsen
cell studies. His seminal work on the uptake and reaction of
trace gases such as ClONO2 and N2O5 with HCl on ice provided
key insights into the role of polar stratospheric clouds in
Antarctic ozone destruction. And, of course, his contributions
to gas-phase kinetics over many decades are legendary. Finally,
Dave has shared most generously his time and expertise with
the kinetics and atmospheric chemistry communities, including
playing a major role in the NASA data evaluations that many
of us in these areas consult on an almost daily basis. We look
forward to many more years of his contributions and exciting
science. Happy Birthday, Dave!
Experimental Section
Combined ATR-Long Path FTIR. Figure 1 is a schematic
diagram of the combined ATR-long path FTIR apparatus. The
shell of the reaction chamber consists of a cylindrical borosili-
cate cell that is 30 cm in diameter and 1.2 m in length. Stainless
steel endplates on which the optics are mounted increase the
total length to 1.4 m, giving a volume of 102 L. A 5.0 cm
diameter quartz tube that houses a lamp is located along the
center of the reaction chamber, which permits irradiation into
the UV beyond the ∼320 nm cutoff for the borosilicate glass
from which the outer chamber was constructed. When the
surface areas of the cell, quartz tube, endplates, and optics are
taken into account, the surface-to-volume ratio is 18.6 m-1. All
exposed metal surfaces inside the reaction chamber are covered
with halocarbon wax (Halocarbon Products, Inc., Series 1500)
to minimize reactivity.
Relative humidity (RH) and temperature (T) are monitored
inside the reaction chamber using a Vaisala gauge (HMP-238).
Pressures in the range from 103 to 0.1 Torr are measured using
a Leybold diaphragm gauge (Ceravac CTR-91), and those in
the range from 750 to 10-4 Torr are measured using a Leybold
Pirani Gauge (Thermovac TTR-216). A BOC-Edwards iQDP80/
iQMB250 dry pump is used to evacuate the cell while
minimizing the backflow of organics associated with oil pumps.
A ThermoNicolet Nexus 670 FTIR spectrometer is used to
record infrared spectra of the gas phase. The IR beam is directed
from the FTIR via a set of transfer optics into the reaction cell
through a ZnSe window. Multiple reflections are achieved using
White cell optics,56 and a total path length of 47.5 m was used
in the present studies. The mirrors in the cell were gold coated
and protected with a thin layer of SiO. The infrared beam exiting
the cell was focused using a 30° off-axis parabolic mirror onto
a liquid nitrogen cooled mercury cadmium telluride (MCT)
detector.
The ATR probe (Figure 2) is a hollow waveguide probe
(Axiom Analytical ATR-FTIR Probe DPR-100) inserted through
a vacuum-tight fitting into the reaction chamber. The ATR-

6888 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Ramazan et al.
relative humidity (RH) was measured in both the chamber and
the mixing bulb. This mixture was allowed to come to
equilibrium for 30 min, and ATR-FTIR and LP-FTIR back-
ground spectra were recorded at 1 cm-1 resolution with 108-
160 coadded scans for the condensed phase and 1 cm-1
resolution and 172 coadded scans for the gas phase. This allowed
most of the water present in the following spectra to ratio out
in the data analysis. Then NO2 was flushed into the cell as a
mixture in nitrogen and the cell was filled to a total pressure of
1 atm with N2, giving ∼87-250 ppm NO2. In earlier studies,26
no difference was observed in the chemistry when the reactions
were carried out in air or N2, so N2 was chosen in order to
minimize NO oxidation. Reaction times up to ∼20 h at a
temperature of 296 ( 1 K were typically used, with spectra
continuously recorded. At the end of the reaction, the contents
of the cell were evacuated for ∼2 h, during which time the cell
pressure dropped to ∼10-2-10-3 Torr. Upon cessation of this
pumping, the gas-phase and surface film spectra were recorded
either under vacuum or after filling the cell to 1 atm with dry
N2.
Figure 2. Schematic of the hollow waveguide ATR probe.
FTIR accessory base is placed into the sample compartment of
a second FTIR spectrometer (ThermoNicolet Avatar 370). The
IR beam in the sample compartment is directed into the infrared
transmitting ATR crystal and undergoes total internal reflection.
The IR beam reflects nine times within the crystal and is then
collected and sent to a cooled MCT detector.
For the photolysis experiments, the chamber was filled to 1
atm with N2, and then irradiated for 4 h with either the low
pressure mercury lamp or the black lamp. The temperature inside
the chamber increased by a maximum of 3 K during the
irradiation.
Two infrared crystals were used during these experiments:
AMTIR (Ge33As12Se55) and silicon (Si). AMTIR has a high IR
throughput and wide spectral range (6000-700 cm-1). The
surface of the silicon crystal has a thin layer of SiOx due to
exposure to air; hence there are some -Si-OH terminal groups
on the surface. This is similar to the surface of the borosilicate
glass that comprises most of the surface in the cell; hence we
assume that ATR measurements of the thin films using a Si
crystal are representative of the majority of the surface chemistry
occurring throughout the cell. However, the silicon crystal does
not transmit below ∼1500 cm-1 due to the oxidized surface
layer. As a result, AMTIR was used in the majority of the studies
reported here and the silicon crystal was used primarily to
validate the observations made over a wider spectral range. In
the spectral region above 1500 cm-1, where measurements could
be made using both crystals, similar results were obtained using
both crystals.
Gas-phase NO2, HONO, and NO were quantified by the net
absorbance of their peaks at 2917, 1263 or 790, and 1875 cm-1
,
respectively. (While there is a much stronger absorption band
for NO2 at 1620 cm-1, there is significant interference from
water vapor that causes additional uncertainty in quantification
using this region.) Detection limits for these gases using these
bands are 2 ppm for NO2 and NO and 0.2 ppm for HONO.
Concentrations of NO2 and NO were determined based on
calibrations using mixtures of known concentrations in N2 in
the cell. The HONO concentrations were calculated using the
1263 or 790 cm-1 band and applying an effective absorption
cross section58 of (3.7 ( 0.4) × 10-19 cm2 molecule-1 (base
10) or (2.8 ( 0.6) × 10-19 cm2 molecule-1 (base 10) to measure
total HONO based on a trans/cis ratio of 2.3.
Chemicals. Nitric oxide (Matheson, 99%) was purified by
passing it through a liquid nitrogen trap to remove impurities
such as NO2 and HNO3. Nitrogen dioxide was synthesized by
reacting the purified NO with excess oxygen (Oxygen Service
Company, 99.993%) for at least 2 h. The NO2 was then
condensed in a coldfinger at 195 K and the excess O2 pumped
off. The nitrogen (Oxygen Service Company, 99.999%), air
(Oxygen Service Company, <0.1 ppm total hydrocarbons, <0.5
ppm CO, <2.0 ppm H2O, <0.5 ppm CO2), and deuterated water
(D2O) (Cambridge Isotope Labs, 99.8%) were used as received.
Cyclohexane (Fisher, 99.9%) and cyclohexanone (Aldrich,
99.8%) were placed in a small glass bulb and underwent three
freeze-pump-thaw cycles to remove any air and any other
low vapor pressure materials in the bulb. The water was
Nanopure ASTM type I reagent grade water (Barnstead, 18.2
MΩ cm).
Theory. To guide the interpretation of the experimentally
obtained spectra, first principles electronic structure calculations
of the magnitude of the shifts in the nitric acid vibrational bands
and their theoretical intensities upon complexation with water
were carried out. Both harmonic analysis and anharmonic
corrections of the fundamentals were carried out for the clusters
as described elsewhere.59,60 The harmonic analysis was per-
formed at the second-order perturbation level of theory (MP2)
using the aug-cc-pVDZ basis set.61,62
Irradiation of the cell contents, both gas phase and surfaces,
is accomplished using lamps located along the axis of the cell.
For the experiments described herein, a low pressure mercury
lamp (GE, 30 W, G30T8) or a black lamp (Sylvania, 30 W,
F30T8/350BL) is used. The low pressure mercury lamp has the
strongest line at 254 nm57 while the black lamp emits a broad
range of light from 300 to 400 nm.
The combined LP-ATR FTIR cell described here differs
from that of Sayer and Horn55 in several important respects.
First, the ATR crystal and gas-phase long path FTIR are co-
located in the same cell, minimizing potential changes in the
composition due to transfer of gases from one chamber to
another. Second, the White cell optics are designed to give long
path lengths (up to ∼100 m) to obtain improved detection for
gaseous species. Third, this system can be used in either the
static or flow mode, while that of Sayer and Horn55 is designed
for use as a flow system. Finally, the cell contents can be
photolyzed using a variety of lamps in order to study photo-
chemical processes.
Experimental Procedures. In a typical experiment, the
reaction chamber was filled with a H2O/N2 mixture at relative
humidities between 37 and 63% to a total pressure of ∼640
Torr by flowing the N2 gas through a bubbler containing water
and mixing it with dry nitrogen in a 5 L mixing bulb. The

Heterogeneous Hydrolysis of NO2
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6889
The anharmonic vibrational spectroscopy calculations were
carried out by the correlation-corrected vibrational self-consistent
field (CC-VSCF) method.63-65 This method treats both the
intrinsic anharmonicity of each mode and the effects due to
anharmonic coupling between different normal modes. The CC-
VSCF approximation can be used directly with potential points
generated from ab initio calculations. In this work, MP2 with a
TZP basis set was used for all cases. The CC-VSCF algorithm
implemented in the GAMESS package of codes66 was em-
ployed. In this variant, only anharmonic couplings between pairs
of normal modes are included, while interactions involving
triplets of normal modes or higher order couplings are neglected.
Evidence from previous calculations shows that CC-VSCF,
including the pairwise mode-mode coupling approximation,
is usually of very good accuracy.65 Part of the results for HNO3,
HNO3-H2O, and HONO were published previously.60 We note
that, for the systems studied here, both the intrinsic anharmo-
nicity of individual modes (e.g., OH stretching modes) and the
anharmonic coupling between different modes proved significant
in cases, depending on the transition. While the effects of
anharmonicity on the absolute vibrational frequencies can be
substantial, the effect is often much smaller when frequency
-
shifts (e.g., between NO3 and NO3--H2O) are considered.
Thus, the anharmonic calculations here were restricted to
molecular species and to their 1:1 complexes with water.
Considerations of frequency shifts were based on the harmonic
calculations only. Finally, the intensities of the vibrational
transitions were also computed anharmonically, using CC-
VSCF.
Results and Discussion
Figure 3. Infrared spectra of water on ATR crystals as a function of
relative humidity for (a) AMTIR and (b) silicon at 296 K.
ATR Measurements of Thin Water Films. Attenuated total
reflectance (ATR) has been widely used for studying thin films
on surfaces.67 In the present studies, it is assumed that the thin
film on the ATR probe is representative of that on the cell walls,
so that it monitors the heterogeneous chemistry occurring on
the cell surfaces in real time. The film is formed by the uptake
of water from the gas phase onto the cell walls and ATR crystal
surface, providing the medium for heterogeneous chemistry.
Upon internal reflection of the infrared beam at the surface
of the crystal, the evanescent wave penetrates the interface and
probes material in contact with the surface. The penetration
depth (dp) of the beam is given by eq I:67
the crystal. The uptake is similar for these two materials,
suggesting that heterogeneous chemistry that is controlled by
water will be similar on AMTIR and silicon, and that the use
of AMTIR with its wider spectral range is appropriate for
following the surface chemistry in these systems. This is also
supported by the similarity of the spectra obtained using the
two crystals in the region above 1500 cm-1 during the
heterogeneous hydrolysis of NO2 (not shown here).
The spectra in Figure 3 at the higher relative humidities are
indistinguishable from those of bulk liquid water. Hence the
liquid water peak can be used to estimate the effective number
of water layers present on the surface as a function of relative
humidity. It is not known whether the water is distributed equally
over the surface as a smooth film or is in islands. The
measurements therefore provide an average, or effective, number
of water layers assuming that it is evenly distributed on the
crystal. With this caveat, the water coverage, Θ, in number of
layers, is calculated from the absorbance spectra using eq II, a
modified form of Beer’s law:68
λ
dp )
(I)
2 1/2
n12π(sin2 θ - n21
)
where λ is the wavelength of the light, n1 is the refractive index
of the denser medium (3.4 for the Si ATR crystal and 2.6 for
the AMTIR crystal), θ is the incident angle, and n21 is the ratio
of the refractive index of the rarer medium (n2) to that of the
crystal (n1). For bulk water in contact with the AMTIR crystal,
for example, the depth of penetration at 3400 cm-1 is 0.37 µm.
However, as shown below, the average thickness of the water
film on this crystal at 50% RH is less than 1 nm. This is
sufficiently thin that the evanescent wave penetrates the entire
film. In this case, the refractive indices of the ATR crystal and
of the media above the film control the electric field more than
the refractive index of the film itself, so that n2 is taken as that
of air. The depth of penetration into air at 3400 cm-1 for the
AMTIR crystal is 0.30 µm.
2.303A
Θ )
(II)
NSH Oσj
2
where A is the base 10 integrated absorbance of the liquid water
peak from 2800 to 4000 cm-1, N is the number of reflections
(nine for this ATR probe), and SH O is the surface density of
2
one water monolayer (1.0 × 1015 molecules cm-2). The
integrated cross section, σj (base e), was calculated for the same
range to be 1.43 × 10-16 cm molecule-1 from optical constants
reported by Downing and Williams.69
Figure 3 shows the absorption spectra in the 2500-4000 cm-1
region for both the AMTIR and silicon crystals as a function
of the relative humidity of the gas phase in equilibrium with
Figure 4 shows the calculated effective number of water layers
on the two crystals as a function of relative humidity. AMTIR

6890 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Ramazan et al.
Figure 6. Typical time series of ATR spectra on an AMTIR crystal
during the NO2 hydrolysis experiment shown in Figure 5. [NO2]o )
184 ppm and 51% RH.
Figure 4. Water uptake isotherms for AMTIR and silicon. Inset is an
enlargement of the region between 0 and 75% RH at 296 K.
cm-1 are assigned to the well-known ν3 asymmetric stretch of
NO3- complexed to water.59,71,72 The magnitude of the splitting
of the degeneracy for this stretch (i.e., the spacing between the
peaks) decreases as the amount of water associated with the
nitrate ion increases. For example, Ritzhaupt and Devlin71
observed a splitting of 150 cm-1 for HNO3 in an argon matrix
containing 6% water compared to 65 cm-1 when the matrix
was 100% water. Theory also predicts this trend, with a
predicted splitting of 101 cm-1 for nitrate associated with four
water molecules and 62 cm-1 for the complex with five water
molecules.59 The splitting of the peaks in Figure 6 is 93 cm-1
,
suggesting that about four to five water molecules are associated
with the nitrate under these conditions.
The asymmetric NO2 stretch in nitric acid (monomer or when
complexed to water or another HNO3) is in the 1650-1710
cm-1 region59,71,73-78 regardless of the environment (gas phase,
low-temperature matrixes, solid phase). We therefore assign the
1710 cm-1 band to nitric acid (without specifying whether it is
the monomer or a complex). Following the experiments of
Barnes et al.77 on nitric acid in low-temperature matrixes, we
assign the 1600 cm-1 band to a nitrogen dioxide-nitric acid
complex.
Figure 5. Concentration-time profiles for a typical NO2 hydrolysis
experiment with 184 ppm NO2 in 1 atm of N2 at 51% RH. Error bars
are (2s.
and silicon are similar in their water uptake at relative humidities
of 80% and below. At ∼50% RH, which was used for most of
the experiments reported here, there is the equivalent of ∼2
layers of water on the crystal surface, similar to that measured
on glass and quartz in earlier studies.70 It is therefore appropriate
to assume that the ATR spectra on both the silicon and AMTIR
crystals are representative of the chemistry that is occurring on
the cell walls.
The band at 2975 cm-1 (Figure 6) is assigned to molecular
nitric acid complexed to water and, as discussed below, to some
extent complexed to HNO3 itself. Support for this assignment
comes from comparison to theoretical calculations described
in more detail below, as well as from several additional
experiments. First, after some time had elapsed to allow the
NO2 hydrolysis to proceed, the cell was pumped out for
approximately 2 h. This removed the gas-phase species as well
as more volatile species on the surface. Figure 7 compares the
ATR spectrum taken at the end of 4 h reaction time to that
after pumping. The peak at ∼2975 cm-1 has blue shifted, by
∼200 cm-1, suggesting that this is associated with some species
that can be converted to a second compound by removing a
more volatile component. Interestingly, the splitting of the peaks
Heterogeneous Hydrolysis of NO2. Since there have been
a number of laboratory studies of the heterogeneous hydrolysis
of NO2 (see discussion in the Introduction), we chose this
reaction to test our new apparatus. Figure 5 shows a plot of the
gas-phase products during a typical experiment at 51% RH. The
studies reported here were all carried out at ∼50% RH in order
to focus on an intermediate water vapor concentration that is
typical of atmospheric conditions, but that is also easily
generated and controlled under laboratory conditions. As
expected, the NO2 concentration decreases, while both HONO
and NO concentrations increase. The sum of the yields of
HONO and NO is constant at 44 ( 19% (2s), consistent with
much of work reviewed in Finlayson-Pitts et al.26 and within
experimental error of 50% expected based on the stoichiometry
of reaction 1. We have proposed a mechanism for this reaction
previously,26,47 which is qualitatively consistent with the
experimental observations; rate constants for the individual steps
are not known at the present time due to the multiple,
interdependent processes in the gas phase and on the surface.
Figure 6 shows the ATR spectra of the surface film during
the same experiment. The absorption bands at 1316 and 1409
-
in the 1400 cm-1 region assigned to NO3 does not change
significantly, suggesting that water is strongly bound to this ion.
However, the shape of the peaks suggests there may be a
contribution from an underlying peak upon pumping, which
could be due to molecular nitric acid,73,75 possibly complexed
to itself or to water.59
Experiments were also carried out in which H2O was replaced
by D2O. Figure 8 compares the ATR spectrum obtained using
D2O to that using H2O. The peak at 2975 cm-1 for the reaction
in H2O has shifted to ∼2275 cm-1, consistent with a deuterium

Heterogeneous Hydrolysis of NO2
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6891
TABLE 1: Summary of Assignment of Infrared Peaks
Observed by ATR-FTIR on the Surface in the NO2
Heterogeneous Hydrolysis at 50% RH at Room Temperature
and Comparison to Literature Values
band positiona
(cm-1
lit. band positions in
assignment
(HNO3)2
)
low-temp matrixesb (cm-1
)
3210
3280 (N2)74
3216 (Ar)77
3235 (Ar)59
HNO3‚H2O
HNO3‚2H2O
HNO3‚3H2O
HNO3‚nH2O
HNO3‚NO2
NO3-‚xH2O
2975
2812
2450
1710
1600
3025 (Ar)59
2788 (Ar)59
2600 (Ar)59
1650-171059,71,72
1611 (Ar)77
1316, 1409
∼1300-140059,71,72
a This study. b The matrix used is in parentheses.
Figure 7. ATR spectra at the end of the NO2 hydrolysis reaction and
after 2 h of pumping. The NO2 hydrolysis spectrum has been multiplied
by 1.82 to normalize it to the pumped-out spectrum in order to match
the integrated area of the 1500-1800 cm-1 region.
frequency after pumping (Figure 7). This should represent the
most dehydrated form of HNO3 on the surface. On the basis of
previous theoretical calculations and experimental studies of
HNO3 in low-temperature inert gas matrixes,59,71,74,77 this band
is assigned to the dimer of nitric acid, (HNO3)2.
On the basis of the peak positions in Figures 6-8, the second
contribution is taken to be at 2975 cm-1, which we assign to
nitric acid bound to one water molecule. This means that
complexation of water to the HNO3 in place of a second HNO3
(i.e., the dimer) that is responsible for the peak at 3210 cm-1
has red shifted the H-ONO2 stretch by ∼235 cm-1. This is in
excellent agreement with the theoretical and low-temperature
matrix results of McCurdy et al.,59 where the shift was observed
experimentally to be 210 cm-1, compared to a calculated shift
of 181 cm-1
.
The third contribution was taken at 2450 cm-1 based on the
obvious shoulder in this region, and in order to match the
experimental data, a fourth peak at 2812 cm-1 was also included.
These peaks are assigned to HNO3(H2O)2 (2812 cm-1) and
HNO3(H2O)3 (2450 cm-1), respectively. While the existence
of the 2812 cm-1 peak is not obvious from the spectra, it is
consistent with experiments and theoretical calculations for
HNO3(H2O)2.59 Furthermore, it would be unusual to have the
monohydrate and the trihydrate, but not the dihydrate.
Figure 8. Comparison of ATR spectra from NO2 hydrolysis using
(black spectrum) H2O and (gray spectrum) D2O. Water has been ratioed
out of both spectra.
isotope effect for an H-O/D-O stretch. The bands in the 1600-
1800 cm-1 region do not shift, nor do those assigned to NO3
,
-
confirming that these are not attributable to vibrations associated
with hydrogen.
Table 1 compares the peak positions observed here to those
from mixtures of HNO3 with water mostly in low-temperature
inert gas matrixes. Given the relatively large width as well as
the overlap of the peaks at room temperature in the present
studies, and that the surface will impose additional constraints
on adsorbed species, the agreement between the band positions
observed here and those reported in the literature is very good.
Figure 9 shows the results of deconvoluting an ATR spectrum
taken during the NO2 heterogeneous hydrolysis into four
contributing peaks at 3210, 2975, 2812, and 2450 cm-1. There
are significant contributions from all four peaks, with that at
2975 cm-1 being the largest. Figure 10 shows a similar
deconvolution for the spectrum obtained after pumping. The
peak at 3210 cm-1 is now the most important contributor, and
the 2975, 2812, and 2450 cm-1 peaks are approximately equal,
suggesting that pumping leads to conversion of the higher
hydrates of HNO3 on the surface into the dehydrated, surface-
bound dimer of nitric acid.
All of these results are consistent with the peak at 2975 cm-1
being due to nitric acid complexed to one or more water
molecules. We designate this species as (HNO3)x(H2O)y, where
x and y represent the numbers of nitric acid and water molecules
in the surface species. There are several lines of evidence to
support this peak assignment. First, theoretical calcula-
tions59,60,78-83 show that complexation of a molecule of nitric
acid in the gas phase to one water molecule or to -SiH3(OH)
or Si(OH)484 shifts the H-ONO2 stretch, which occurs at 3550
cm-1 in the gas phase, by approximately 500 cm-1. Complexing
nitric acid to two or more water molecules further red shifts
this vibration.59,78 Second, the shift from 2975 to 2275 cm-1
when D2O is used in place of H2O is consistent with a deuterium
isotope effect for the H-ONO2 stretch. Finally, the blue shift
of the peak upon pumping suggests that a volatile species, likely
water, is being selectively removed from these complexes.
While the 2975 cm-1 band is very broad, which is expected
for a hydrogen-bonded H-O stretching vibration at room
temperature, the shapes of the spectra in Figures 6-8 suggest
that this band is comprised of contributions from several species.
Deconvolution of the broad peak between 2000 and 3700 cm-1
into contributions from four species was accomplished using
the following approach. The first contribution was taken to be
at 3210 cm-1, based on the blue shift of the peak to this
-
Changes in the relative intensities of the peaks due to NO3
and the nitric acid complexes can be used to obtain some insights
into the shifts between the various hydrates and ions on
pumping. Table 2 summarizes the peak positions and assign-
ments and the experimentally measured integrated band intensi-
ties for these species in a typical experiment. The relative
concentrations shown in the last column taking the nitrate ion

6892 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Ramazan et al.
TABLE 2: Estimated Concentrations of Nitric Acid
Complexes Relative to Nitrate Ions on the Surface during a
Typical NO2 Heterogeneous Hydrolysis Experiment and
after Pumping
experimental
peak
integrated
ratio of
position
peak
assignment
band intensity
complex concn
(cm-1
)
(cm-1
)
to nitrate ion concna
During Hydrolysis
1316, 1409 NO3-‚xH2O
0.90
0.46
0.66
0.25
0.39
(1.0)b
0.15c
0.39d
0.10e
0.14f
3210
2975
2812
2450
(HNO3)2
HNO3‚H2O
HNO3‚2H2O
HNO3‚3H2O
After Pumping
1316, 1409 NO3-‚nH2O
0.625
2.3
(1.0)b,g
1.1c
3210
2975
2812
2450
(HNO3)2
Figure 9. Deconvolution of the ATR absorbance band (Figure 7)
between 4000 and 2000 cm-1 at the end of NO2 hydrolysis. The
irregular black line represents the spectrum, the dotted black line is
the baseline, the smooth black lines are the individual peaks that make
up the spectrum, and the solid line through the data is a summation of
the individual contributions.
HNO3‚H2O
HNO3‚2H2O
HNO3‚3H2O
0.25
0.32
0.27
0.21d
0.18e
0.13f
a These were obtained from the ratio of (absorbance/theoretical
intensity) for each peak to the corresponding ratio for the (1316 +
1409 cm-1) peaks that were assigned to the nitrate ion-water complex.
b The theoretical intensity for the nitrate ion was estimated as described
in the text to be 592 km mol-1 c A theoretical intensity of 2043 km
.
mol-1 was calculated by adjusting the harmonic value of 1821 km mol-1
upward by 12%; this is the amount by which the harmonic and
anharmonic calculations differ for the complex of HNO3 with one water
molecule, for which both types of calculations were carried out. d A
theoretical intensity of 1121 km mol-1 from the anharmonic calculations
was used for the complex with one water molecule. This is 12% larger
than the value of 999 km mol-1 from the harmonic model. e A
theoretical intensity of 1625 km mol-1 was used by adjusting the
harmonic value of 1448 km mol-1 upward by 12%; this is the amount
by which the harmonic and anharmonic calculations differ for the
complex of HNO3 with one water molecule, for which both types of
f
calculations were carried out. A theoretical intensity of 1896 km mol-1
was used based by adjusting the harmonic value of 1690 km mol-1
upward by 12%; this is the amount by which the harmonic and
anharmonic calculations differ for the complex of HNO3 with one water
molecule, for which both types of calculations were carried out. g On
pumping, the nitrate-water peaks decreased to 70% of the intensity at
the end of the hydrolysis just prior to pumping. However, for ease of
comparison, the relative concentrations of the nitric acid-water
complexes are referenced to the nitrate peaks under the same conditions.
Figure 10. Deconvolution of the broad absorbance (Figure 7) between
4000 and 2000 cm-1 after pumping for 2 h. The irregular black line
represents the spectrum, the dotted black line is the baseline, the smooth
black lines are the individual peaks that make up the spectrum, and
the solid line through the data is a summation of the individual
contributions.
harmonic calculations. For consistency, the harmonic values of
the theoretical intensities for the higher hydrates and the dimer
were therefore adjusted upward by 12%.
-
as the reference, Nx/NNO , were obtained from a Beer-Lambert
3
law type of calculation:
There are a number of caveats to be applied to these estimates.
First, the calculated theoretical intensities are for gas-phase
species, whereas the experimental data are for surface-adsorbed
molecules. In addition to the differences that are present for
gas versus condensed phase, there is also the possibility of
specific interactions with the surface of the crystal in the
experiments, for example with surface hydroxyl groups. Second,
some of the nitrate may be associated directly with the crystal
components rather than only with surface water. Third, there is
a possibility of unrecognized complexes contributing to some
of the peaks; for example, Thi et al.85 have reported several
different complexes of nitrate ions with nitric acid in thin films.
Finally, the amount of material on the crystal, and the corre-
sponding signals, is quite small, so that the estimates of relative
concentrations of surface species should be taken as semiquan-
titative.
Nx
Ax/Ix
)
(III)
N
A
-/I
-
-
NO3
NO3 NO3
-
where Ax and ANO are the experimental integrated band
intensities and Ix and INO are the calculated theoretical
3
-
3
intensities. The theoretical intensity for the nitrate ion was
estimated as follows. A value of 845 km mol-1 was calculated
-
using the anharmonic approach for NO3 complexed to one
water molecule. Harmonic calculations predict that the theoreti-
cal intensity for n ) 4 water molecules is 62% of that for n )
1, while for n ) 5 water molecules the theoretical intensity is
78% of that for n ) 1. Assuming that the 845 km mol-1 can be
scaled accordingly, and that four to five water molecules are
attached to the NO3-, an average theoretical intensity for the
nitrate complexed to water of 592 km mol-1 was derived.
Theoretical intensities are available from the harmonic calcula-
tions for all of the species identified here, and from the
anharmonic calculations for the monohydrate. The theoretical
intensity for HNO3(H2O) from the anharmonic calculations is
1121 km mol-1, 12% larger than the 999 km mol-1 from the
However, within these caveats, some conclusions can be
drawn from the application of the theoretical assignments of
species and band intensities. First, during the NO2 heterogeneous
hydrolysis at 50% RH, the concentrations of the nitric acid-
water complexes are significant relative to the nitrate ion,

Heterogeneous Hydrolysis of NO2
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6893
varying from 10 to 39% of the nitrate ion concentration. In
addition, small amounts of HNO3 dimer are present. On
pumping, there is a significant shift in the surface species toward
the less hydrated forms of nitric acid, including the nitric acid
dimer which becomes the most intense peak (Figure 10).
Although nitric acid is a strong acid, it requires association with
-
at least four water molecules86-88 to dissociate to H+ and NO3
in the gas phase. If pumping removes water preferentially, then
dissociation will be impaired and the nitrate ion will be
converted first to molecular nitric acid and its complexes with
water, and finally to the dimer.
In short, the data are consistent with the formation of both
dissociated and undissociated HNO3 on the surface during the
NO2 heterogeneous hydrolysis, with the nitrate ion and HNO3
both being complexed to water. Equilibria such as the following
are likely involved:
Figure 11. Concentration-time profiles for HONO (a) from irradiation
at 254 nm in the absence (O) and presence (0) of an OH scavenger
(C6H12); (b, 9) from exposure to N2 at 30% RH without irradiation;
(c, b) from exposure to N2 at 50% RH; (d, 4) from exposure to N2 at
50% RH with irradiation at 254 nm beginning at 2000 s; (e, 4) from
irradiation using the black lamp (300-400 nm). Each experiment was
performed after 20 h of NO2 hydrolysis followed by pumping for 2 h.
All data have been averaged over three data points. Error bars shown
in the x direction account for the data averaging; error bars in the y
direction represent the 2s experimental and quantification error.
HNO3 + HNO3 T (HNO3)2
HNO3 + H2O T HNO3‚H2O
(2)
(3)
(4)
(5)
HNO3‚H2O + H2O T HNO3‚2H2O
HNO3‚2H2O + H2O T HNO3‚3H2O
HNO3‚3H2O + xH2O T H3O+ + NO3 ‚yH2O
(6)
-
band and hence its peak height, as well as the relative theoretical
intensities/absorption cross sections.
When the system is pumped out, the concentrations shift in
favor of molecular nitric acid complexed to fewer water
molecules and ultimately to the dimer. Thus the extent of
ionization on the surface is a sensitive function of the amount
of available water. The data in Table 2 can be used to calculate
the ratio of the concentration of the nitrate ions using the bands
at 1316 and 1409 cm-1 to the undissociated nitric acid
complexes with water and with a second HNO3. This gives a
ratio of nitrate ions to the complexes of 1.3:1 during the
hydrolysis at 50% RH, where there is an effective two layers
of water on the surface initially. This suggests that the degree
of dissociation of HNO3 under these conditions is ∼56%. After
pumping, the ratio of nitrate ions to the complexes is 0.63:1,
corresponding to a degree of dissociation of 39%. This is not
surprising, since the degree of dissociation in concentrated bulk
aqueous solutions of nitric acid is also small. For example, the
degree of dissociation in bulk aqueous solutions varies from
50% at 10 M to only 1% at 22 M.89 Further evidence for the
relatively small extent of ionization of the acid comes from the
lack of an underlying broad continuum above 2000 cm-1 which
is seen in amorphous nitric acid-ice films75,90 and which has
been attributed to proton transfer to form a Zundel-type cation.91
While quantitative agreement would not be expected between
solution-phase data and thin surface films which likely have a
more two-dimensional character, the comparison is qualitatively
consistent with increased dehydration of the surface film on
pumping.
Nitric acid has been observed previously37,38 as a product on
high surface area silica during the heterogeneous hydrolysis of
NO2, based on infrared bands in the 1600-1700 cm-1 (asym-
metric -NO2 stretch) region. Infrared spectroscopy of silica
surfaces treated with pure gaseous nitric acid also suggested
that HNO3 formed surface complexes with water based on
absorptions in the H-O stretch region around 3000 cm-1 26,39,40
.
The preferential removal of water from these nitric acid treated
surfaces by pumping was observed.26 Sayer and Horn55 reported
that exposure of solid NaNO3 on an ATR crystal to NO2 gave
nitrate and molecular nitric acid, which was attributed to the
reaction of NO2 with small amounts of water on the NaNO3.
They observed the loss of a band at 3400 cm-1 due to the
adsorbed water and an increase in a peak at ∼3000 cm-1 which
they assigned to H3O+. Our studies suggest the latter is due
rather to a nitric acid-water complex.
In summary, the studies reported here not only conclusively
establish the formation of such complexes during the NO2
heterogeneous hydrolysis, but also show the progression from
ions to increasingly dehydrated forms of nitric acid bound to
the surface as water is removed during pumping.
Photolysis of Surface-Adsorbed Species. Given the pro-
posed photolysis of adsorbed nitric acid as a source of HONO
in the troposphere,44,46 a set of experiments was carried out in
which the cell surfaces were photolyzed after carrying out the
hydrolysis and pumping, followed by filling the chamber to 1
atm with N2. From the interpretation of the data discussed above,
the cell walls should hold adsorbed nitric acid and nitrate under
these conditions (Figures 7 and 10). Two lamps were used
during separate experiments: (1) a low pressure mercury lamp
(primarily 254 nm, with smaller peaks into the UV/visible57)
and (2) a black lamp (300-400 nm).
Reaction 6 implies that hydronium ion should be formed in
concentrations equal to those of nitrate ions. The bending mode
for H3O+ in nitric acid trihydrate in ice has been reported75,76,92
to be in the region around 1750 cm-1. The theoretical intensity
for this H3O+ band has been calculated64 to be 139 km mol-1
,
about a factor of 4 less than that calculated for nitrate complexed
to water (592 km mol-1; see footnote b to Table 2). A weak
shoulder can be seen on the high wavenumber side of the 1710
cm-1 peak in Figures 6-8, but its intensity is quite small. This
may reflect differences between conditions in the gas phase and
those in the surface film, which may affect the width of the
Irradiation using the low pressure mercury lamp generated
both HONO and NO in the gas phase. Production of NO and
HONO from the photolysis of nitric acid in a low-temperature
nitrogen matrix was also reported by Chen et al.93 Figures 11
and 12 show the concentration-time profiles for HONO and
NO, respectively, during irradiation. Because both NO and

6894 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Ramazan et al.
HO-NOx photochemistry, some experiments were carried out
in which the chamber was heated by ∼3 °C using heating tape.
There was no detectable HONO or NO generated in the gas
phase. After 4 h of heating, ∼100 ppb gaseous HNO3 was
measured. From a mass balance perspective, desorption of HNO3
followed by its photolysis is therefore unlikely to be the source
of NO and HONO since much larger amounts of NO + HONO,
∼3 ppm, are formed in the same time period during irradiation.
Irradiation using the black lamps generated only NO, and at
a rate approximately a factor of 4 less than that at 254 nm. The
lack of observation of HONO may be due to the smaller light
intensities at longer wavelengths and/or to the strong absorption
of light in the 300-400 nm region by HONO.97 The photolysis
rate constant for HONO using the black lamps was measured
to be kp(300-400 nm) ) (2.0 ( 0.4) × 10-4 s-1 (2s). Again,
the lifetime for HONO photolysis in the cell, τ ) 5 × 103 s, is
too long to explain the observed NO formation via HONO
photolysis. Thus, the difference in photolysis lifetimes using
the two lamps is less than a factor of 2 and is unlikely to explain
the lack of observation of HONO using the black lamps.
Zhou et al.46 reported HONO and NOx (mainly NO2)
production when a glass surface that had been exposed to
gaseous HNO3 and water vapor was photolyzed using a medium
pressure mercury lamp with a cutoff filter that only transmitted
above 290 nm. In contrast, we observe HONO formation only
when 254 nm light is used. However, their sensitivity for HONO
is much greater than that using FTIR in the present studies, so
that our data do not unequivocally rule out HONO formation
during 300-400 nm irradiation. In addition, it is not clear
whether the composition of the surface complexes is different
when formed from the NO2 heterogeneous hydrolysis as in our
experiments compared to saturating the surface with gas-phase
nitric acid and water as in the Zhou et al. experiments.46
Thermodynamically, there are a number of potential overall
reactions that could generate NO and HONO from the photolysis
of surface-adsorbed nitric acid-water complexes. For example,
reaction 7
Figure 12. Concentration-time profiles for NO (a) from irradiation
at 254 nm in the absence (O) and presence (0) of an OH scavenger
(C6H12); (b, 9) from exposure to N2 at 30% RH without irradiation;
(c, b) from exposure to N2 at 50% RH; (d, b) from exposure to N2 at
50% RH with irradiation at 254 nm beginning at 2000 s; (e, 4) from
irradiation using the black lamp (300-400 nm). Each experiment was
performed after 20 h of NO2 hydrolysis followed by pumping for 2 h.
All data have been averaged over three data points. Error bars shown
in the x direction account for the data averaging; error bars in the y
direction represent the 2s experimental and quantification error.
HONO are generated, it is possible that NO is the primary
product and that HONO is formed from the recombination of
NO with OH radicals. The latter could ostensibly be formed
from some process such as nitrate ion photolysis94 to O-, which
then reacts with residual water to form OH. To test this
hypothesis, cyclohexane was added as an OH scavenger in some
runs. The reaction between cyclohexane and OH is sufficiently
fast (k ) 7.2 × 10-12 cm3 molecule-1 s-1 at 298 K)95 that this
should have scavenged >95% of any OH that is generated. The
open squares in Figures 11 and 12 represent the products in the
presence of cyclohexane and the open circles those without
cyclohexane; there is no significant impact on the production
of HONO (or NO), ruling out the reaction of NO with OH as
the source of HONO.
HNO3 + H2O + hν f HONO + H2O2
(7)
The possibility that NO is generated by HONO photolysis
was also considered. The photolysis rate constant for HONO
in the cell was measured using a technique described else-
where,47 and was found for the low pressure mercury lamp to
be kp ) (1.2 ( 0.4) × 10-4 s-1 (2s), corresponding to a lifetime
of τ ) 8.3 × 103 s. This is too slow to produce the NO shown
in Figure 12, which increases immediately upon irradiation.
To assess whether increased amounts of water would enhance
the HONO and NO production, some runs were carried out in
which humidified N2 was added to the cell after pumping.
Previous studies96 established that borosilicate glass cells that
had been exposed to gas-phase HONO and then evacuated
became a source of HONO upon exposure to water vapor.
Similarly, when water was added to the cell in the present studies
after reaction followed by pump down, an increase in gas-phase
HONO was also observed (filled squares in Figure 11). When
the lamp is turned on during such a run at 50% RH, no
enhancement in the HONO production over that due to
desorption from the cell walls is observed (filled circles in Figure
11). Similarly, as seen in Figure 12, there is no enhancement
in the rate of NO production during irradiation in the presence
of water vapor.
is energetically possible in the gas phase at wavelengths below
741 nm. However, given that there is likely a barrier involved
in such a reaction, the actual wavelength of light that would
effect such a change is likely significantly shorter. The same
applies to reactions 8 and 9 below. Of course, the reaction
enthalpy will also be modified by the surface, but this calculation
suggests that the low pressure mercury lamps and black lamps
both should have sufficient energy to drive such photochemistry.
Similarly, reaction 8 is possible at wavelengths below 393 nm:
HNO3 (+H2O) + hν f HONO (+H2O) + O
(8)
Generation of NO could occur in a reaction such as
HNO3 (+H2O) + hν f NO (+H2O) + HO2
(9)
which is energetically accessible at wavelengths below 501 nm,
if there is no additional barrier to the dissociation.
Clearly, further work will be required to assess the photo-
chemistry of the surface-adsorbed products of the NO2 hetero-
geneous hydrolysis.
When the lamps are turned on, the temperature of the cell
rises by a maximum of 3 °C. To test whether HONO and NO
could be generated simply by increased thermal desorption, or
by desorption of HNO3 followed by its photolysis to initiate
Atmospheric Implications. There is increasing evidence that
processes occurring on surfaces in the tropospheric boundary
layer significantly impact the chemistry and photochemistry in
this region. Figure 13 shows the amount of water on the ATR

Heterogeneous Hydrolysis of NO2
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6895
peak at 1600 cm-1 was assigned. Whether the absorption
spectrum of the dimer shifts substantially on complexation to a
second HNO3, to NO2, or possibly to NO3- as observed by Thi
et al.85 is not known, but is the subject of ongoing work.
It was observed here as well as in previous studies96 that gas-
phase HONO is generated when a glass surface that had been
exposed to gaseous HONO was pumped out and water vapor
then added; in contrast, the addition of dry N2 does not result
in gas-phase HONO. Thus either HONO or a HONO precursor
must be present on the surface and interact with water to form
and/or desorb HONO. This process is very important for the
interpretation of HONO concentrations in air and their depen-
dence on RH.36 Although we did not observe bands in the ATR
spectra that were assignable to adsorbed HONO, if it were
complexed to nitrate ions or other surface species, the spectra
could be shifted significantly compared to that expected based
on gas-phase data.
Figure 13. Water uptake on some common atmospheric surfaces (from
ref 31) and on AMTIR and silicon ATR crystals.
Diamond and co-workers2-4 have shown that organic com-
pounds are adsorbed on surfaces in the boundary layer, and that
the composition of this layer changes from urban to downwind
rural sites. Oxidation of organics in air is well-known,1 and the
oxidation products would be expected at least in part to be taken
up onto surfaces after formation in the gas phase. The present
results raise the possibility that the oxidation may actually occur
on the surface itself. In the case of surface films from the NO2
heterogeneous hydrolysis, the formation of oxygenated and
nitrated products could be a marker for such surface oxidations
in the boundary layer.
Finally, airborne dust particles are known to become distrib-
uted globally,100-104 impacting global chemistry and climate.
Given that dust generally has a high silica content, the chemistry
of the oxides of nitrogen described here is expected to occur
not only on urban surfaces but also on such dust particles during
transport. There is some evidence for HONO production from
reactions on dust particles. For example, Stutz and co-workers35
reported enhanced HONO concentrations during a dust storm
in Phoenix, AZ.
crystals used here, in comparison to materials present in the
boundary layer such as soil and vegetation. Clearly, the amount
of water on the surfaces in our experiments is comparable to
that on materials found close to the Earth’s surface; hence the
results reported here should be applicable to chemistry in the
boundary layer.
Heterogeneous reactions of oxides of nitrogen, which are
precursors to ozone, nitric acid, and airborne particles,1 are
particularly important from local to global scales. Nitric acid is
an “end product” of this chemistry, and has been generally
assumed to dissociate into nitrate and hydronium ions on
surfaces. The work presented here suggests that a significant
fraction of nitric acid may exist as complexes with water on
atmospheric surfaces such as building materials, foliage, and
dust particles, and that photolysis can regenerate gas-phase NOx
(“renoxification”). Thus, formation of nitric acid on surfaces
or uptake from the gas phase may not permanently remove NOx
from the atmosphere, as is often assumed.
In addition, oxides of nitrogen, including nitric acid, are
associated with significant health effects.98 Whether the form
of nitric acid (i.e., molecular or nitrate ions) plays a role in
determining the health impacts of acids and particles is not
known, but it is clearly of interest given that much of the acid
on the surface of dust particles, for example, will be undisso-
ciated molecular HNO3 complexed to water.
These studies are the first to report the simultaneous measure-
ment of both gases and surface species during the heterogeneous
hydrolysis of NO2. While a great deal of insight into the nature
of the reaction, its intermediates, and products has been gained,
further studies are clearly needed to identify the specific surface
species responsible for the production of HONO from surfaces
upon exposure to water vapor and upon irradiation.
It has been well-documented that there are species of
unknown identity on the walls of environmental chambers that
generate gas-phase HONO upon irradiation of supposedly
“clean” chambers.49-54 One possibility is that it is the nitric
acid-water surface complexes observed here that photolyze to
generate HONO. This would be consistent with the experiments
of Zhou and co-workers in which HONO production was
observed upon photoysis of nitric acid treated glass46 or of a
contaminated sampling system.44 However, Rohrer et al.54
reported that HONO generation from the walls of the SAPHIR
chamber continued (albeit at a reduced rate) when a filter that
only transmitted light above 370 nm was used. The absorption
cross sections for gas-phase HNO3 fall off rapidly above 300
nm,97 and theoretical studies99 show that binding of water to
nitric acid in the gas phase does not red shift the absorption
spectrum significantly. The surface would have to cause very
large and unexpected shifts in the electronic absorption spectrum
of nitric acid and its water complexes if it were to be the wall
source of HONO. However, as discussed above, one of the peaks
observed in the ATR spectra is attributed to the nitric acid dimer.
Another possibility is the NO2-HNO3 complex to which the
Acknowledgment. We are especially grateful to Jorg Meyer
and Lee Moritz, without whose technical assistance with the
design and construction this apparatus would not have been
possible. We are also grateful to the California Air Resources
Board (Contract No. 00-323) and the National Science Founda-
tion (ATM-0097573, Collaborative Research in Chemistry CHE-
0209719 and Environmental Molecular Sciences Institute CHE-
0431512) for support of this work. Battelle operates the Pacific
Northwest National Laboratory for the Department of Energy.
This research was performed in part using the Molecular Science
Computing Facility (MSCF) in the William R. Wiley Environ-
mental Molecular Sciences Laboratory, a national scientific user
facility sponsored by the Department of Energy’s Office of
Biological and Environmental Research located at Pacific
Northwest National Laboratory.
References and Notes
(1) Finlayson-Pitts, B. J.; Pitts, J. N. Chemistry of the Upper and Lower
Atmosphere: Theory, Experiments and Applications; Academic Press: San
Diego, 2000.

6896 J. Phys. Chem. A, Vol. 110, No. 21, 2006
Ramazan et al.
(2) Diamond, M. L.; Gingrich, S. E.; Fertuck, K.; McCarry, B. E.; Stern,
G. A.; Billeck, B.; Grift, B.; Brooker, D.; Yager, T. D. EnViron. Sci. Technol.
2000, 34, 2900.
(3) Gingrich, S. E.; Diamond, M. L.; Stern, G. A.; McCarry, B. E.
EnViron. Sci. Technol. 2001, 35, 4031.
(4) Lam, B.; Diamond, M. L.; Simpson, A. J.; Makar, P. A.; Truong,
J.; Hernandez-Martinez, N. A. Atmos. EnViron. 2005, 39, 6578.
(5) Wayne, L. G.; Yost, D. M. J. Chem. Phys. 1951, 19, 41.
(6) Cathala, J.; Weinrich, G. Compt. Rend. 1952, 244, 1502.
(7) Peters, M. S.; Holman, J. L. Ind. Eng. Chem. 1955, 47, 2536.
(8) Goyer, G. G. J. Colloid Sci. 1963, 18, 616.
(9) Graham, R. F.; Tyler, B. J. J. Chem. Soc., Faraday Trans. 1 1972,
68, 683.
(10) England, C.; Corcoran, W. H. Ind. Eng. Chem. Fundam. 1974, 13,
373.
(11) Ten Brink, H. M.; Bontje, J. A.; Spoelstra, H.; van de Vate, J. F.
Atmospheric Pollution 1978. In Studies in EnVironmental Science; Benarie,
M. M., Ed.; Elsevier: Amsterdam, 1978; Vol. 1; p 239.
(12) Sakamaki, F.; Hatakeyama, S.; Akimoto, H. Int. J. Chem. Kinet.
1983, 15, 1013.
(13) Pitts, J. N.; Sanhueza, E.; Atkinson, R.; Carter, W. P. L.; Winer,
A. M.; Harris, G. W.; Plum, C. N. Int. J. Chem. Kinet. 1984, 16, 919.
(14) Svensson, R.; Ljungstrom, E.; Lindqvist, O. Atmos. EnViron. 1987,
21, 1529.
(15) Akimoto, H.; Takagi, H.; Sakamaki, F. Int. J. Chem. Kinet. 1987,
19, 539.
(44) Zhou, X.; Hi, Y.; Huang, G.; Thornberry, T. D.; Carroll, M. A.;
Bertman, S. B. Geophys. Res. Lett. 2002, 29, DOI: 10.1029/2002GL015080.
(45) Zhou, X.; Civerolo, K.; Dai, H.; Huang, G.; Schwab, J.; Demerjian,
K. J. Geophys. Res. 2002, 107, DOI: 10.1029/2001JD001539.
(46) Zhou, X. L.; Gao, H. L.; He, Y.; Huang, G.; Bertman, S. B.;
Civerolo, K.; Schwab, J. Geophys. Res. Lett. 2003, 30, 2217.
(47) Ramazan, K. A.; Syomin, D.; Finlayson-Pitts, B. J. Phys. Chem.
Chem. Phys. 2004, 6, 3836.
(48) Kleffmann, J.; Gavriloaiei, T.; Hofzumahaus, A.; Holland, F.;
Koppmann, R.; Rupp, L.; Schlosser, E.; Siese, M.; Wahner, A. Geophys.
Res. Lett. 2005, 32.
(49) Carter, W. P. L.; Atkinson, R.; Winer, A. M.; Pitts, J. N. J. Int. J.
Chem. Kinet. 1981, 13, 735.
(50) Carter, W. P. L.; Atkinson, R.; Winer, A. M.; Pitts, J. N. J. Int. J.
Chem. Kinet. 1982, 14, 1071.
(51) Carter, W. P. L.; Atkinson, R.; Winer, A. M.; Pitts, J. N. Atmos.
EnViron. 1985, 19, 1977.
(52) Sakamaki, F.; Akimoto, H. Int. J. Chem. Kinet. 1988, 20, 111.
(53) Killus, J. P.; Whitten, G. Z. Int. J. Chem. Kinet. 1990, 22, 547.
(54) Rohrer, F.; Bohn, B.; Brauers, T.; Bru¨ning, D.; Johnen, F.-J.;
Wahner, A.; Kleffman, J. Atmos. Chem. Phys. Discuss. 2004, 4, 7881.
(55) Sayer, R. M.; Horn, A. B. Phys. Chem. Chem. Phys. 2003, 5, 5229.
(56) White, J. U. J. Opt. Soc. Am. 1942, 32, 285.
(57) Calvert, J. G.; Pitts, J. N., Jr. Photochemistry; John Wiley and
Sons: New York, 1966.
(58) Barney, W. S.; Wingen, L. M.; Lakin, M. J.; Brauers, T.; Stutz, J.;
Finlayson-Pitts, B. J. J. Phys. Chem. A 2000, 104, 1692; J. Phys. Chem. A
2001, 105, 4166.
(59) McCurdy, P. R.; Hess, W. P.; Xantheas, S. S. J. Phys. Chem. A
2002, 106, 7628.
(60) Miller, Y.; Chaban, G. M.; Gerber, R. B. Chem. Phys. 2005, 313,
213.
(61) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007.
(62) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. J. Chem. Phys.
1992, 96, 6796.
(16) Jenkin, M. E.; Cox, R. A.; Williams, D. J. Atmos. EnViron. 1988,
22, 487.
(17) Perrino, C.; DeSantis, F.; Febo, A. Atmos. EnViron. 1988, 22, 1925.
(18) Febo, A.; Perrino, C. Atmos. EnViron. 1991, 25A, 1055.
(19) Bambauer, A.; Brantner, B.; Paige, M.; Novakov, T. Atmos.
EnViron. 1994, 28, 3225.
(20) Wiesen, P.; Kleffmann, J.; Kurtenbach, R.; Becker, K. H. Faraday
Discuss. 1995, 100, 121.
(21) Kleffmann, J.; Becker, K. H.; Wiesen, P. Atmos. EnViron. 1998,
32, 2721.
(22) Kleffmann, J.; Becker, K. H.; Wiesen, P. J. Chem. Soc., Faraday
Trans. 1998, 94, 3289.
(23) Harrison, R. M.; Collins, G. M. J. Atmos. Chem. 1998, 30, 397.
(24) Kurtenbach, R.; Becker, K. H.; Gomes, J. A. G.; Kleffmann, J.;
Lo¨rzer, J. C.; Spittler, M.; Wiesen, P.; Ackermann, R.; Geyer, A.; Platt, U.
Atmos. EnViron. 2001, 35, 3385.
(25) Kotamarthi, V. R.; Gaffney, J. S.; Marley, N. A.; Doskey, P. V.
Atmos. EnViron. 2001, 35, 4489.
(26) Finlayson-Pitts, B. J.; Wingen, L. M.; Sumner, A. L.; Syomin, D.;
Ramazan, K. A. Phys. Chem. Chem. Phys. 2003, 5, 223.
(27) Platt, U.; Perner, D.; Harris, G. W.; Winer, A. M.; Pitts, J. N., Jr.
Nature 1980, 285, 312.
(28) Winer, A. M.; Biermann, H. W. Res. Chem. Intermed. 1994, 20,
423.
(29) Calvert, J. G.; Yarwood, G.; Dunker, A. M. Res. Chem. Intermed.
1994, 20, 463.
(30) Lammel, G.; Cape, J. N. Chem. Soc. ReV. 1996, 25, 361.
(31) Lammel, G. Formation of Nitrous Acid: Parameterization and
Comparison with ObserVations; Max-Planck-Institut-fu¨r Meteorologie:
Mainz, 1999.
(32) Alicke, B.; Platt, U.; Stutz, J. J. Geophys. Res. 2002, 107, DOI:
10.1029/2000JD000075.
(33) Stutz, J.; Alicke, B.; Neftel, A. J. Geophys. Res. 2002, 107, DOI:
10.1029/2001JD000390.
(34) Alicke, B.; Geyer, A.; Hofzumahaus, A.; Holland, F.; Konrad, S.;
Patz, H.; Schafer, J.; Stutz, J.; Volz-Thomas, A.; Platt, U. J. Geophys. Res
2003, 108, DOI: 10.1029/2001JD000579.
(35) Wang, S. H.; Ackermann, R.; Spicer, C. W.; Fast, J. D.; Schmeling,
M.; Stutz, J. Geophys. Res. Lett. 2003, 30, doi: 10.1029/2003GL017014.
(36) Stutz, J.; Alicke, B.; Ackermann, R.; Geyer, A.; Wang, S. H.; White,
A. B.; Williams, E. J.; Spicer, C. W.; Fast, J. D. J. Geophys. Res. 2004,
109; doi: 10.1029/2003JD004135.
(37) Goodman, A. L.; Underwood, G. M.; Grassian, V. H. J. Phys. Chem.
A 1999, 103, 7217.
(38) Barney, W. S.; Finlayson-Pitts, B. J. J. Phys. Chem. A 2000, 104,
171.
(39) Rivera-Figueroa, A. M.; Sumner, A. L.; Finlayson-Pitts, B. J.
EnViron. Sci. Technol. 2003, 37, 548.
(40) Dubowski, Y.; Sumner, A. L.; Menke, E. J.; Gaspar, D. J.; Newberg,
J. T.; Hoffman, R. C.; Penner, R. M.; Hemminger, J. C.; Finlayson-Pitts,
B. J. Phys. Chem. Chem. Phys. 2004, 6, 3879.
(41) Kleffmann, J.; Benter, T.; Wiesen, P. J. Phys. Chem. A 2004, 108,
5793.
(63) Chaban, G. M.; Jung, J. O.; Gerber, R. B. J. Chem. Phys. 1999,
111, 1823.
(64) Chaban, G. M.; Jung, J. O.; Gerber, R. B. J. Phys. Chem. A 2000,
104, 2772.
(65) Gerber, R. B.; Chaban, G. M.; Brauer, B.; Miller, Y. First Principles
Calculations of Anharmonic Vibrational Spectroscopy of Large Molecules.
In Theory and Applications of Computational Chemistry: The First 40
Years; Dykstra, C. E., Frenking, G., Kim, K. S., Scuseria, G. E., Eds.;
Elsevier: Amsterdam, 2005; Chapter 9, p 165.
(66) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;
Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.;
Su, S. J.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem.
1993, 14, 1347; http://www.msg.ameslab.gov/GAMESS/GAMESS.html.
(67) Harrick, N. J. Internal Reflection Spectroscopy; Interscience
Publishers: New York, 1967.
(68) Foster, M. C.; Ewing, G. E. J. Chem. Phys. 2000, 112, 6817.
(69) Downing, H. D.; Williams, D. J. Geophys. Res. 1975, 80, 1656.
(70) Sumner, A. L.; Menke, E. J.; Dubowski, Y.; Newberg, J. T.; Penner,
R. M.; Hemminger, J. C.; Wingen, L. M.; Brauers, T.; Finlayson-Pitts, B.
J. Phys. Chem. Chem. Phys. 2004, 6, 604.
(71) Ritzhaupt, G.; Devlin, J. P. J. Phys. Chem. 1977, 81, 521.
(72) Sporleder, D.; Ewing, G. E. J. Phys. Chem. A 2001, 105, 1838.
(73) McGraw, G. E.; Bernitt, D. L.; Hisatsune, I. C. J. Chem. Phys.
1965, 42, 237.
(74) Guillory, W. A.; Bernstein, M. L. J. Chem. Phys. 1975, 62, 1058.
(75) Ritzhaupt, G.; Devlin, J. P. J. Phys. Chem. 1991, 95, 90.
(76) Smith, R. H.; Leu, M.-T.; Keiser, L. F. J. Phys. Chem. 1991, 95,
5924.
(77) Barnes, A. J.; Lasson, E.; Nielsen, C. J. J. Mol. Struct. 1994, 322,
165.
(78) Escribano, R.; Couceiro, M.; Go´mez, P. C.; Carrasco, E.; Moreno,
M. A.; Herrero, V. J. J. Phys. Chem. A 2003, 107, 651.
(79) Tao, F.-M.; Higgins, K.; Klemperer, W.; Nelson, D. D. Geophys.
Res. Lett. 1996, 23, 1797.
(80) Staikova, M.; Donaldson, D. J. Phys. Chem. Chem. Phys. 2001, 3,
1999.
(81) Kjaergaard, H. G. J. Phys. Chem. A 2002, 106, 2979.
(82) Ferna´ndez, D.; Botella, V.; Herrero, V. J.; Escribano, R. J. Phys.
Chem. 2003, 107, 10608.
(83) Dimitrova, Y. Spectrochim. Acta A 2004, 60, 1.
(84) Thompson, K. C.; Margey, P. Phys. Chem. Chem. Phys. 2003, 5,
2970.
(85) Thi, M. P.; Herzog-Cance, M. H.; Potier, A.; Potier, J. J. Raman
Spectrosc. 1981, 11, 96.
(42) Mochida, M.; Finlayson-Pitts, B. J. J. Phys. Chem. A 2000, 104,
9705.
(43) Saliba, N.; Yang, H.; Finlayson-Pitts, B. J. J. Phys. Chem. A 2001,
105, 10339.
(86) Kay, B. D.; Hermann, V.; Castleman, A. W. J. Chem. Phys. Lett.
1981, 80, 469.

Heterogeneous Hydrolysis of NO2
J. Phys. Chem. A, Vol. 110, No. 21, 2006 6897
(87) Zhang, X.; Mereand, E.; Castleman, A. W. J. J. Phys. Chem. 1994,
98, 3554.
(88) Gilligan, J. J.; Castleman, A. W. J. J. Phys. Chem. A 2001, 105,
5601.
(89) Davis, W. J.; DeBruin, H. J. J. Inorg. Nucl. Chem. 1964, 26, 1069.
(90) Tisdale, R. T.; Prenni, A. J.; Iraci, L. T.; Tolbert, M. A. Geophys.
Res. Lett. 1999, 26, 707.
(91) Zundel, Z. In The Hydrogen Bond, Recent DeVelopments in Theory
and Experiments; Schuster, P., Zundel, Z., Sandorfy, C., Eds.; North-
Holland: Amsterdam, 1976; Vol. 2, Chapter 15.
(92) Tolbert, M. A.; Middlebrook, A. M. J. Geophys. Res. 1990, 95,
22.
(93) Chen, W.-J.; Lo, W.-J.; Cheng, B.-M.; Lee, Y.-P. J. Chem. Phys.
1992, 97, 7167.
(94) Mack, J.; Bolton, J. R. J. Photochem. Photobiol., A 1999, 128, 1.
(95) Aschmann, S. M.; Atkinson, R. J. Phys. Chem. A 1997, 101, 8042.
(96) Syomin, D. A.; Finlayson-Pitts, B. J. Phys. Chem. Chem. Phys.
2003, 5, 5236.
(97) Sander, S. P.; Friedl, R. R.; Golden, D. M.; Kurylo, M. J.; Huie,
R. E.; Orkin, V. L.; Moorgat, G. K.; Ravishankara, A. R.; Kolb, C. E.;
Molina, M. J.; Finlayson-Pitts, B. J. Chemical Kinetics and Photochemical
Data for Use in Atmospheric Studies. EValuation No. 14; JPL Publ. No.
02-25; Jet Propulsion Laboratory: Pasadena, CA, 2002.
(98) Gauderman, W. J.; Avol, E.; Gilliland, F.; Vora, H.; Thomas, D.;
Berhane, K.; McConnell, R.; Kuenzli, N.; Lurmann, F.; Rappaport, E.;
Margolis, H.; Bates, D.; Peters, J. New Engl. J. Med. 2004, 351, 1057.
(99) Brown, E.; Gerber, R. B. Unpublished data, 2005.
(100) Prospero, J. M.; Nees, R. T. Science 1977, 196, 1196.
(101) Prospero, J. M.; Glaccum, R. A.; Nees, R. T. Nature 1981, 289,
570.
(102) Prospero, J. M.; Nees, R. T. Nature 1986, 320, 735.
(103) Prospero, J. M.; Schmitt, R.; Cuevas, E.; Savoie, D. L.; Graustein,
W. C.; Turekian, K. K.; Volz-Thomas, A.; Diaz, A.; Oltmans, S. J.; Levy,
H. Geophys. Res. Lett. 1995, 22, 2925.
(104) Prospero, J. M. J. Geophys. Res. 1999, 104, 15917.