A study of the heterogeneous reaction between dinitrogen pentaoxide and chloride ions on low-temperature thin films

Article abstract of DOI:10.1021/jp993763l

When low-temperature thin films of either ionic or covalent dinitrogen pentaoxide, N₂O_5, are exposed to gaseous HCl and water, the only products observed in the solid phase by reflection - absorption infrared spectroscopy (RAIRS) are molecular nitric acid and the oxonium ion. Nitryl chloride, ClNO_2, is not detectable. When dinitrogen pentaoxide is co-deposited with hydrogen chloride and water at 85 K and annealed to 140 K, the resultant RAIR spectra indicate that the film is composed of H₃O⁺Cl^-, N₂O₅, HNO_3, and D_2h-N₂O₄. When nitryl chloride is co-deposited with either water or HCl/water mixtures, infrared spectra indicative of solid D_2h-N₂O₄ are measured, as well as peaks corresponding to nitrate ions and cis-ClONO (chlorine nitrite). Reaction between ClNO2 and its isomer, cis-ClONO, is proposed as an explanation for the formation of dinitrogen tetraoxide in both systems. The proposed reaction mechanism for this hydrolysis is extended to the N₂O₅/H₂O/HCl deposits in order to explain the lack of observable ClNO₂ in such thin films.

Full text of DOI:10.1021/jp993763l

1890
J. Phys. Chem. A 2000, 104, 1890-1897
A Study of the Heterogeneous Reaction between Dinitrogen Pentaoxide and Chloride Ions
on Low-Temperature Thin Films
John R. Sodeau,* Tristan B. Roddis, and Matt P. Gane
Department of Chemistry, UniVersity College Cork, Cork, Ireland
ReceiVed: October 21, 1999; In Final Form: December 14, 1999
When low-temperature thin films of either ionic or covalent dinitrogen pentaoxide, N2O5, are exposed to
gaseous HCl and water, the only products observed in the solid phase by reflection-absorption infrared
spectroscopy (RAIRS) are molecular nitric acid and the oxonium ion. Nitryl chloride, ClNO2, is not detectable.
When dinitrogen pentaoxide is co-deposited with hydrogen chloride and water at 85 K and annealed to 140
K, the resultant RAIR spectra indicate that the film is composed of H3O+Cl-, N2O5, HNO3, and D2h-N2O4.
When nitryl chloride is co-deposited with either water or HCl/water mixtures, infrared spectra indicative of
solid D2h-N2O4 are measured, as well as peaks corresponding to nitrate ions and cis-ClONO (chlorine nitrite).
Reaction between ClNO2 and its isomer, cis-ClONO, is proposed as an explanation for the formation of
dinitrogen tetraoxide in both systems. The proposed reaction mechanism for this hydrolysis is extended to
the N2O5/H2O/HCl deposits in order to explain the lack of observable ClNO2 in such thin films.
Introduction
All of these reactions are significant in atmospheric chemistry
as they serve to convert inert solid or aqueous chloride to gas-
phase “active” chlorine compounds. In particular, reactions 1
and 4 produce the highly photolabile gas-phase compounds
ClNO and ClNO2 that subsequently go on to form Cl and NOx
radicals when exposed to solar radiation.10,11
The product halogen radicals can go on to abstract hydrogen
atoms from hydrocarbons and hence initiate the chain reactions
that lead to photochemical smog. Indeed, it has been suggested
that Cl radicals are responsible for 20-40% of the nonmethane
hydrocarbon oxidation in marine environments.12
The most well-studied of the above four reactions is the one
involving the important trace species, dinitrogen pentaoxide,
which acts as a temporary reservoir for nitrogen oxides. This
surface process is important because the product, nitryl chloride,
(ClNO2) is sufficiently reactive to form secondary products,
which are potentially relevant to atmospheric chemical composi-
tion. For example, experimental studies show that ClNO2 reacts
with aromatic organic compounds known to be present in the
troposphere. Specifically, flow tube observations indicate that
the nitration of phenols is observed after exposure to gaseous
ClNO2.6 Furthermore, environmental chamber measurements
have shown that chlorinated organic compounds are formed
upon exposure to ClNO2.13 It can also react heterogeneously
with ionic anions, including bromide,14 iodide,15 and nitrite14
to form a variety of gas-phase halogenated products.
The NOx radicals produced as a consequence of ClNO2
formation can go on to take part in catalytic chain reactions,
which cause a net increase in tropospheric ozone. Reactions 1
and 4 are especially important in this respect because they lead
to a regeneration of gas-phase NOx that would otherwise have
been removed from the atmosphere by the deposition of N2O5.
As mentioned above, nitryl chloride is also extremely photo-
labile: it dissociates in a facile manner when exposed to visible
light below 870 nm11,16 as indicated by reaction 5
The chemical composition of the troposphere is influenced
by reactions occurring in both the gas phase and at the interfaces
between the gas, solid, and/or liquid phase. Compared to
homogeneous chemical processing, the role of heterogeneous
chemistry within the atmosphere is not well understood. This
problem is due to the compounded difficulty of both monitoring
and categorizing the huge variety of interfacial surfaces that
are present and then predicting the reactions that can take place
on/within them.1
One of the primary effects of heterogeneous chemistry on
the atmosphere is the removal of trace constituents by interaction
with particles and aerosols. For example, the adsorption of nitric
acid on tropospheric surfaces leads to a reduction of daytime
NO2 concentration. However, it is the heterogeneous potential
for the formation of unexpected gaseous compounds from
surfaces that is currently least understood in models of
atmospheric composition. For instance, a number of reactions
are known to occur between atmospheric trace gases and sea
salt aerosols but their kinetomechanistic pathways are not
understood.2-6 The most important contributing surface reactions
are those involving chloride ions or HCl, and so the current
state of knowledge for the chemistry associated with the
processes is outlined below.
A deficiency of chloride ions on particles collected in or near
urban areas compared to those found in seawater has often been
observed.7-9 This reduction in Cl- concentrations has been
attributed to a number of ion-exchange reactions between
atmospheric trace gases and the halide surface as follows:
N2O5 (g) + NaCl (s or aq) f
ClNO2 (g) + NaNO3 (s or aq) (1)
HNO3 (g) + NaCl (s or aq) f
HCl (g) + NaNO3 (s or aq) (2)
H2SO4 (g) + 2NaCl (s or aq) f
2HCl (g) + Na2SO4 (s or aq) (3)
ClNO2 (g) + hν (λ < 870 nm) f Cl (g) + NO2 (g) (5)
Reaction 1 can occur with chloride ions that are either in a solid
form5,17 or in aqueous solution.6,15 In fact, droplet-train measure-
ments have established that the reactive uptake of N2O5 occurs
2NO2 (g) + NaCl (s or aq) f
ClNO (g) + NaNO3 (s or aq) (4)
10.1021/jp993763l CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/15/2000

Reaction of Dinitrogen Pentaoxide and Chloride
J. Phys. Chem. A, Vol. 104, No. 9, 2000 1891
more rapidly on sodium chloride solution than on pure water.18
The reaction is of primary importance in the marine troposphere
due to the high concentration of sea-salt, but it has also been
proposed that it may occur in the stratosphere after volcanic
eruptions.19 Such a hypothesis is borne out by the observation
of NaCl particles in the lower stratosphere after the El Chichon
volcanic eruption in 1982; this event was accompanied by an
increase in the HCl column density of around 40% relative to
background levels.20,21 Thus, reaction 1 can also produce active
chlorine species in the lower stratosphere if sodium chloride
particles are present, although it will be competing with the
direct production of hydrogen chloride due to their reaction with
gas-phase nitric acid22 as follows:
coupled to a Digilab FTS-60A spectrometer with aluminum
mirror transfer optics. Vacuum-compatible KBr windows were
used in conjunction with mirrors to focus the IR beam onto the
sample position at a grazing angle of incidence (ca. 75°). The
reflected beam was detected by a HgCdTe detector cooled to
77 K. The polycrystalline gold foil used as a substrate was
mounted between a pair of tungsten supports which were in
thermal contact with a liquid nitrogen reservoir. The substrate
temperature was measured using a chromel/alumel thermocouple
directly spot-welded to the gold foil; it could be varied between
80 and 1000 K with a stability of (0.5 K. Inside the chamber,
the gases were directed onto the substrate by glass guidance
tubes. The total exposure of the substrate was measured by a
cold cathode ionization gauge (MKS).
Ultrathin films of dinitrogen pentaoxide and hydrogen
chloride hydrates (H3O+Cl-‚nH2O) were produced by effusive
deposition on to the cold gold substrate. N2O5 was synthesized/
purified according to the method of Davidson et al.31 No
impurities were observable in its gas-phase IR spectrum. The
water was purified prior to deposition; HCl gas was obtained
from BDH (99.6% purity), and no impurities were observed in
either its gas-phase infrared or mass spectra. The compounds
were either simultaneously deposited via separate glass dosing
lines, or the acid mixtures were deposited onto a prepared N2O5
surface. The mixtures were monitored where possible using
quadrupole mass spectrometry and reaction sequences were
followed in the temperature range 85-170 K.
Nitryl chloride was prepared from the room-temperature
reaction between 50 mbar N2O5 and 100 g dry sodium chloride
in a darkened bulb to prevent photolysis of the product.5 The
ClNO2 was transferred to a blackened glass reservoir and was
found to be stable for 24 h. The main decomposition (impurity)
products are N2O4 and ClNO as evidenced in the IR spectra as
will be discussed. Hence, freshly prepared samples of ClNO2
were deposited on to the cold gold substrate either in pure form
or with water or with HCl/water mixtures.
Dinitrogen tetroxide was prepared by the deposition of
gaseous NO2 (BDH, 99.5%) onto the cold gold-foil surface. No
impurities were detectable from the RAIR spectra obtained.
In all cases, RAIR spectra of the resulting submicrometer
films were recorded at a resolution of 4 cm-1 by the co-addition
of 256 double-sided interferograms.
HNO3 (g) + NaCl (s or aq) f
HCl (g) + NaNO3 (s or aq) (2)
The reaction between hydrogen chloride and dinitrogen pen-
taoxide occurs in a manner analogous to the reaction of N2O5
with a chloride ion surface:
N2O5 (g) + HCl (s or aq) f
ClNO2 (g) + HNO3 (s or aq) (6)
In the stratosphere, reaction 6 does not occur to any significant
degree in the gas-phase.23 However, if polar stratospheric clouds
(PSC’s) are present, this heterogeneous reaction would lead to
a reactivation of chlorine from the HCl reservoir species.24 The
process also effectively couples the odd-nitrogen/odd-chlorine
catalytic cycles and can lead to denitrification of the stratosphere
by the production of solid-phase nitric acid.25
The important, competing reaction for N2O5 in both the
troposphere and stratosphere is the following heterogeneous
hydrolysis reaction:
N2O5 (g) + H2O (s or l) f 2HNO3 (s or aq)
(7)
Numerous experimental measurements of this reaction have been
reported, involving a variety of techniques including flow-tube
studies,26 droplet-train,18,27 and water clusters.28 RAIR spectro-
scopic investigations of the solid-state interaction between co-
deposited N2O5 and water-ice show that no reaction occurs at
temperatures between 85 and 160 K, even after holding the film
for a period of several hours. However, when N2O5 is deposited
onto a solid water-ice film at 140 K, products from the gas-
solid reaction are observed, although not when water is deposited
onto frozen N2O5.29,30 In the stratosphere, reaction 7 leads to
denitrification, which in turn affects the ambient ozone level
as less gas-phase NOx becomes available to sequester active
chlorine.
As shown above, the occurrence and balance between
reactions 1, 6, and 7 have potentially important impacts on
“active” chlorine release into the troposphere. The surface
changes in terms of reaction intermediates, associated with these
reactions have not been monitored directly before. Therefore,
the purpose of the current experiments is to observe the effects
of the interactions in N2O5/HCl/chloride ion/H2O low-temper-
ature thin films of varying composition using reflection-
absorption infrared spectroscopy (RAIRS). By this means, it is
hoped to ascertain whether the main implication drawn from
kinetics investigations, i.e., that ClNO2 is the primary product
of the surface reaction between N2O5 and chloride ions, is
always valid for the system.
Results
RAIR Spectroscopy of N2O5/HCl/H2O Low-Temperature
Films. Figure 1 shows the RAIR absorbance spectra produced
before (a) and after (b) dosing HCl (ca. 5 × 10-7 mbar) and a
small amount of water (ca.1 × 10-8 mbar) on top of N2O5 films
grown and maintained at 135 K. This temperature was chosen
because in this regime N2O5 exists in its ionic form, i.e.,
nitronium nitrate, NO2+NO3-. Some water was included in the
dosing mixture to promote the probability of adsorption, because
molecular HCl alone will not stick to the substrate, as has been
discussed previously.32 The difference spectrum between Figure
1a and Figure 1b is shown as Figure 1c. Loss peaks are observed
-
for the NO2+NO3 reactant, while simultaneous growth peaks
at 1670, 1310, and 956 cm-1 are observed, which can be
assigned to molecular nitric acid by comparison both with
literature values and an authentic sample shown as Figure 2b.33,34
The peaks correspond to the antisymmetric NO2 stretch, the
symmetric NO2 stretch, and the NO stretching mode, respec-
tively. The broad features observed in parts b and c of Figure
1 centered on 1700 and 1200 cm-1 are due to the formation of
H3O+ ions.35 A clear and direct comparison between the
Experimental Section
The RAIR spectroscopy experiments were performed utilizing
a UHV stainless steel vacuum chamber, which was optically

1892 J. Phys. Chem. A, Vol. 104, No. 9, 2000
Sodeau et al.
Figure 1. Reaction of a thin film of nitrosonium nitrate with HCl at 135 K: (a) RAIR spectrum of a thin film of NO2+NO3-; (b) RAIR spectrum
-
of thin film of NO2+NO3 following dosing of HCl/H2O; (c) RAIR difference spectrum of spectra b and a.
NO2+NO3-/HCl spectrum and the HNO3 spectrum obtained in
the RAIRS apparatus is shown as parts a and b of Figure 2.
Identical results were obtained when the experiments were
performed on surfaces held at 85 K, i.e., when N2O5 is present
exclusively in its amorphous covalent form.29
Thus, it appears that when HCl/H2O is dosed on to either
amorphous, covalent, or ionic dinitrogen pentaoxide, the only
identifiable solid-phase products are molecular nitric acid along
with H3O+ ions. For both the ionic and covalent starting
materials, reaction stops once a layer of molecular acid has built
up thereby “poisoning” the surface.
Upon comparison of the above measurements with previously
published work, no IR absorption bands due to nitryl chloride
appear to be formed in our experiments at either temperature
condition. This fact is surprising in view of the well-documented
reaction 6 discussed above,
strong 810 cm-1 absorption band for ClNO2 is clearly absent
in Figure 2a. Comparison could not be made at 135 K because
nitryl chloride desorbs before this temperature is reached. The
implications are that, in the reaction experiments, the ClNO2
product has either desorbed from the substrate surface or
undergone further reaction. Although the former is likely in the
case of reaction with nitronium nitrate at 135 K, it should not
occur in the case of covalent nitrogen pentaoxide maintained
at 85 K.
Complementary experiments in which N2O5, HCl, and H2O
were simultaneously deposited via separate glass dosing lines
onto the cold substrate were also performed. Figure 3 shows
the resulting RAIR spectrum after deposition at 85 K, and then
after annealing the mixed film to a variety of temperatures.
The spectrum of the initial film held at 85 K (Figure 3a) is
complicated. It indicates that a small amount of reaction has
occurred because it comprises mainly a mixture of amorphous
covalent N2O5, amorphous HCl monohydrate, molecular nitric
acid, plus some of the so-called “low-temperature” form of
dinitrogen tetraoxide previously discussed elsewhere.36 Upon
annealing the film to 140 K (Figure 3b), bands due to amorphous
N2O5 disappear, while a new set appears at 1766, 1740, 1271,
764, and 741 cm-1. These features correspond to the high-
N2O5 (g) + HCl (s or aq) f
ClNO2 (g) + HNO3 (s or aq) (6)
To confirm this fact, an authentic sample of ClNO2 was prepared
and a fresh deposit made on to the gold foil surface held at 80
K. The comparison is shown as parts a and c of Figure 2; the

Reaction of Dinitrogen Pentaoxide and Chloride
J. Phys. Chem. A, Vol. 104, No. 9, 2000 1893
Figure 2. Comparison spectra for nitrosonium nitrate HCl reactions in a frozen thin film: (a) RAIR spectrum of thin film of NO2+NO3- following
reaction with HCl at 135 K; (b) RAIR spectrum of thin film of HNO3 at 85 K; (c) RAIR spectrum of thin film of ClNO2 at 85 K.
temperature form of amorphous/crystalline D2h-dinitrogen
tetraoxide.36-39 For comparison, an authentic sample of N2O4
was prepared by the deposition of NO2 onto the cold gold-foil
surface; the resulting spectrum is shown as Figure 4. In Figure
3b, a small peak at 800 cm-1 also appears, which may
correspond to the antisymmetric NO2 deformation mode of
ClNO2 or of the nitrite ion.
Upon further warming to 160 K (Figure 3c), the majority of
the D2h-N2O4 desorbs, along with the HCl, as evidenced by the
loss of broad peaks corresponding to the hydrated oxonium ion
at 2580, 2125, 1720, and 1140 cm-1. Only molecular nitric acid
and water remain, along with some amorphous H3O+NO3-, as
revealed by the increased intensity of the nitrate antisymmetric
stretching mode at 1445 cm-1. Finally, heating to 170 K (Figure
3D) results in the desorption of much of the molecular nitric
acid and water, accompanied by the formation of peaks due to
crystalline nitric acid dihydrate.35 At no point was nitronium
nitrate observed.
min dosing of a freshly prepared sample of ClNO2 at a pressure
of 1 × 10-7 mbar on to the clean gold substrate held at 85 K.
Absorption peaks at ∼1660, 1274, and ∼810 cm-1 correspond
to the ν4 antisymmetric stretch, ν1 symmetric stretch, and ν2
deformation modes of the NO2 group, respectively. The absorp-
tion at 1323 cm-1 is due to the overtone band 2ν6, which has
high intensity due to Fermi resonance with the ν1 stretching
mode.40 Modes are blue-shifted from the values obtained using
transmission infrared spectroscopy due to the observation of
longitudinal optical modes rather than transverse optical modes,
as expected for RAIR spectra. Finally, small peaks observed at
ca.1750 and 1930 cm-1 are due to N2O4 and ClNO impurities,
respectively. From IR analysis of authentic samples of these
materials on low-temperature surfaces, the levels of impurity
typically present in the main experiments are assessed as <1%.
The above assignments are summarized in Table 1.
Figure 6 shows the result of co-depositing ClNO2 at a pressure
of 1 × 10-7 mbar with water at a pressure of 1 × 10-6 mbar
for 30 s. By comparison with Figure 5 (pure nitryl chloride), it
is clear that a new strong peak has appeared at 1743 cm-1. This
feature is accompanied by an apparent increase in intensity of
the ν1 modes of ClNO2, which have shifted slightly to 1271
cm-1. Furthermore, the ν2 mode at 810 cm-1 broadens, increases
in intensity, and grows a “sideband” feature centered at 760
The production of large amounts of D2h dinitrogen tetraoxide
in the spectra displayed in Figure 3, as well as the fact that no
nitryl chloride product is observed are surprising features.
Therefore, a further set of experiments involving pure ClNO2
were undertaken in order to clarify the mechanism.
RAIR Spectroscopy of ClNO2/H2O Low-Temperature
Films. Figure 5 shows the RAIR spectrum produced after 3

1894 J. Phys. Chem. A, Vol. 104, No. 9, 2000
Sodeau et al.
Figure 3. RAIR spectra of (a) mixed N2O5/HCl/H2O film at 80 K, then annealed to (b) 140 K, (c) 160, or (d) 170 K.
TABLE 1: Observed Infrared Frequencies (cm-1) and
Vibrational Assignments for Solid Nitryl Chloride
as shown in Figure 6 is observed when ClNO2 and HCl are
co-deposited with water.
Effusive deposition of a water layer on top of solid ClNO2
maintained at 85 K showed no reactant loss or N2O4 formation.
normal mode
RAIRSa
transmissionb
ν4 (NO2 asym stretch)
1675 m, sh
1648 s
1323 s
1657 s
1633 s
1320 s
1316 s
1261 s
1256 s
804 m
786 m
654 sh
652 m
2ν6
Discussion
ν1 (NO2 sym stretch)
ν2 (NO2 bend)
ν6(NO2 out-of-pl bend)
1274 s
Isomerization of ClNO2 to ClONO on a Low-Temperature
Surface. The reaction between dinitrogen pentaoxide and
hydrogen chloride is expected to produce both nitric acid and
nitryl chloride as discussed above.
811 m
791 m, sh
653 w
a This work, Figure 5. b From ref 40.
N2O5 (g) + HCl (s or aq) f
cm-1. As shown in Figure 4, a spectrum with concurrently
growing absorbances at 1740, 1270, and 760 cm-1 corresponds
to the production of amorphous D2h-N2O4. The ν4 mode of
ClNO2 is still observable as a single absorption centered around
1665 cm-1. A broad peak at ca. 850 cm-1 is also present due
to the lattice mode of amorphous water-ice. Small peaks at
1925 and 1400 cm-1 correspond to nitrosyl chloride and nitrate
impurities, respectively.
ClNO2 (g) + HNO3 (s or aq) (6)
From parts b and c of Figure 2, it can be seen that the infrared
frequencies of nitryl chloride overlap to an extent with those
of molecular nitric acid. However, if ClNO2 is present, it should
be readily identifiable by the appearance of its ν2 NO2 bending
mode at around 810 cm-1. This feature is not observed in the
spectra displayed as parts b and c of Figure 1. Hence, only the
nitric acid product of reaction 6 is identified in the solid phase.
Its origin is not due to a simple stoichiometric reaction with
water as given in reaction 7 because it has been shown
previously that N2O5 does not hydrolyze when water is deposited
The most important feature is a new, sharp peak at 1717 cm-1
,
which is most readily assigned to the strong ν1 NdO stretching
mode of the cis isomer of chlorine nitrite, ClONO as measured
by three other research groups.41-43 The same product spectrum

Reaction of Dinitrogen Pentaoxide and Chloride
J. Phys. Chem. A, Vol. 104, No. 9, 2000 1895
Figure 4. RAIR spectrum of D2h-N2O4 formed after the deposition of gaseous NO2/N2O4 on to the gold substrate held at 120 K.
Figure 5. RAIR spectrum of pure nitryl chloride at 85 K.
onto it at any temperature between 80 and 160 K.29,30
ions present. In the simple two-component system, there are
only limited possibilities for the hydrolysis of nitryl chloride.
Room-temperature flow-tube experiments by Behnke et al.6 led
them to postulate that the reaction between nitryl chloride and
water proceeds via the dissociated, hydrated ionic intermediate
NO2+Cl-. The overall reaction was described as follows:
N2O5 (g) + H2O (s or l) f 2HNO3 (s or aq)
(7)
The following questions therefore arise from the thin-film
observations. (1) What secondary reactions of nitryl chloride
can occur in the system? (2) How is the dinitrogen tetraoxide
formed? (3) Are the first two questions related?
ClNO2 + 3H2O f 2H3O+ + NO3- + Cl-
(8)
Production of the dinitrogen tetraoxide occurs when nitryl
chloride and water are co-deposited with or without chloride
+
However, the well-known absorption spectrum due to the NO2

1896 J. Phys. Chem. A, Vol. 104, No. 9, 2000
Sodeau et al.
Figure 6. RAIR spectrum of nitryl chloride co-deposited with water at 85 K.
cation was not observed in any of our low-temperature studies
of nitryl chloride. Hence, at the very least, it can be concluded
that product channel 8 is not dominant in a thin-film.
One stoichiometric reaction that would account for the
observed formation of the N2O4 is the following heterogeneous
degradation reaction:44
empirical calculations using the Hyperchem software package
(Autodesk Inc., release 3) indicate that while ClNO2 has a
negative charge of -0.031 associated with the chlorine atom,
the cis-ClONO isomer has a substantial Clδ+ charge of +0.103.
Reaction 10 may then take place via a concerted reaction
analogous to the low-temperature hydrolysis of chlorine nitrate
discussed in previously published studies.46
Low-Temperature Surface Mechanism for the Reaction
between N2O5 and HCl. The surprising lack of observable
ClNO2 in the sequential deposition spectra measured for HCl
onto a dinitrogen pentaoxide film (Figure 1) can be readily
explained in terms of a mechanism, already published, for the
exposure of frozen dinitrogen pentoxide to acidic water.29 It
can be summarized by the following sequence:
2ClNO2 f N2O4 + Cl2
(9)
It is clear from Figure 5 that this reaction does not occur at 85
K as no products are observed; in fact, no such chemistry is
observed at any temperature before desorption takes place.
An examination of the enthalpy changes associated with
reaction 9 reveals that the reaction is only slightly thermody-
namically favorable with a reaction enthalpy of -6 kJ mol-1 45
.
However, the overall reaction enthalpy decreases to -48 kJ
mol-1 if an alternative stoichiometric reaction is considered in
which one partner is the less stable isomer, ClONO.42
N2O5 + H3O+ f NO2+ + H2O + HNO3
NO2+ + 2H2O f HNO3 + H3O+
(11)
(12)
ClONO + ClNO2 f N2O4 + Cl2
(10)
The production of dinitrogen tetraoxide in the presence of a
large amount of water within co-deposited, mixed N2O5/H2O/
HCl films (Figure 3) can be explained in relation to the
mechanism by which hydrogen chloride reacts with chlorine
nitrate on a water-ice surface.46 Here, the process of reactant
surface coadsorption promotes the formation of partially ionized
intermediates. Hence if, on a surface, the relatively large chloride
ion reactant attacks the more accessible oxygen atoms of
dinitrogen pentaoxide rather than either of the two sterically
protected nitrogen atoms, then the initial product of the reaction
will be the reactive cis-ClONO species rather than ClNO2. The
following mechanism would than operate in addition to (11)
and (12).
This mechanism would represent a plausible explanation for
the observed production of N2O4 in mixed water/ClNO2 films,
as long as it can be shown that the act of surface co-deposition
leads to some isomerization from ClNO2 to ClONO. The RAIR
results do indeed provide evidence for this process because an
absorption band at 1717 cm-1 appears, which is most readily
assigned to the strong ν1 NdO stretching mode of the cis isomer
of nitryl chloride (ClONO), as previously measured in three
other studies.41-43 The same peak is observed for the co-
deposition of ClNO2 with HCl and water at 85 K.
The theory is further borne out by a consideration of the
electron distribution for the two isomers. Simple AM1 semi-

Reaction of Dinitrogen Pentaoxide and Chloride
J. Phys. Chem. A, Vol. 104, No. 9, 2000 1897
N2O5 + H3O+Cl- f δ+ClONO + H3O+NO3
(13)
(14)
(15)
-
(6) (a) Behnke, W.; Elend, M.; Frenzel, A.; Kruger, H. U.; Scheer,
V.; Sikorski, R.; Zetzsch, C. Proceedings of the 4th International Conference
on Chemical Kinetics; NIST: Gaithersburg, MD, 1997. (b) Behnke, W.;
George, C.; Scheer, V.; Zetzsch, C. J Geophys. Res. D 1997, 102, 3795.
(7) Robbins, R. C.; Cadle, R. D.; Eckhardt, D. L. J. Met. 1959 16, 53.
(8) Duce, R. A.; Zoller, W. H.; Moyers, J. L. J Geophys. Res. 1973,
78, 7802.
(9) Clegg, D. L.; Brimblecombe, P. EOS 1986, 67, 877.
(10) Grimley, A. J.; Houston, P. L. J. Chem. Phys. 1980, 72, 1471.
(11) Nelson, H. H.; Johnston, H. S. J Phys. Chem. 1981, 85, 3891.
(12) Singh, H. B.; Kasting, J. F. J. Atmos. Chem. 1988, 7, 261.
(13) Zetzsch, C.; Behnke, W. Ber. Bunsen-Ges. Phys. Chem. 1992, 96,
488.
-
δ+ClONO + H3O+Cl- f Cl2 + H3O+NO2
H3O+NO2- + N2O5 f N2O4 + H3O+NO3
The net result being
2N2O5 + 2H3O+Cl- f N2O4 + Cl2 + 2H3O+NO3
-
-
(16)
(14) Frenzel, A.; Scheer, V.; Sikorski, R.; George, C.; Behnke, W.;
Zetzsch, C. J. Phys. Chem. A 1998, 102, 1329.
(15) George, C.; Behnke, W.; Scheer, V.; Zetzsch, C.; Magi, L.; Ponche,
J. L.; Mirabel, P. Geophys. Res. Lett. 1995, 22, 1505.
(16) Illies A. J.; Takacs, G. A. J. Photochem. 1976, 6, 35.
(17) Finlayson-Pitts, B. J.; Ezell, M. J.; Pitts, J. N., Jr. Nature 1989,
337, 241.
The reaction between nitrite ions and dinitrogen pentaoxide to
form N2O4, i.e., step 15 is well-known and proceeds with a
reaction enthalpy of -123 kJ mol-1 45
.
Conclusions
(18) George, C.; Ponche, J. L.; Mirabel, P.; Behnke, W.; Scheer, V.;
Zetzsch, C. J Phys. Chem. 1994, 98, 8780.
(19) Michelangeli, D. V.; Allen, M.; Yung, Y. L. Geophys. Res. Lett.
1991, 18, 673.
(20) Woods, D. C.; Chuan, R. L.; Rose W. I. Science 1985, 230, 170.
(21) Manki, W. G.; Coffey, M. T. Science. 1984, 226,170.
(22) Leu, M-T.; Timonen, R. S.; Keyser, L. F. J. Phys. Chem. 1995,
99, 13203.
(23) Cantrell, C. A.; Davidson, J. A.; Shetter, R. E.; Anderson, B. A.;
Calvert, J. G. J. Phys. Chem. 1987, 91, 6017.
(24) Tolbert, M. A.; Rossi, M. J.; Golden, D. M. Science 1988, 240,
1018.
(25) Wofsy, S. C.; Molina, M. J.; Salawitch, R. A.; Fox, L. E.; McElroy,
M. B. J. Geophys. Res. 1988, 93-Dl, 2442.
(26) Msibi, I. M.; Li, Y.; Shi, P.; Harrison, R. M. J. Atmos. Chem. 1994,
18, 291.
(27) Van Doren, J. M.; Watson, L. R.; Davidovits, P.; Worsnop, D. R.;
Zahniser, M. S.; Kolb, C. E. J. Phys. Chem. 1990, 94, 3265.
(28) Ahmed, M.; Apps, C. J.; Buesnel, R.; Hughes, C.; Hillier, I. H.;
Watt, N. E.; Whitehead, J. C. J. Phys. Chem. A 1997, 101, 1254.
(29) Horn, A. B.; Koch, T. G.; Chesters, M. A.; McCoustra, M. R. S.;
Sodeau, J. R. J. Phys. Chem. 1994, 98, 946.
(30) Koch, T. G.; Banham, S. F.; Sodeau, J. R.; Horn, A. B.; McCoustra,
M. R. S.; Chesters, M. A. J. Geophys. Res. 1997, 102-D1, 1513.
(31) Davidson J. A.; Viggiano, A. A.; Howard, C. J.; Fehsenfeld, F. C.;
Albritton, D. L.; Ferguson, E. E. J. Chem. Phys. 1978, 68, 2085.
(32) Banham, S. F.; Sodeau, J. R.; Horn, A. B.; McCoustra, M. R. S.;
Chesters, M. A. J. Vac. Sci. Technol. A 1996, 14, 1620.
(33) Koch, T. G.; Sodeau, J. R. J. Chem. Soc., Faraday Trans. 1996,
92, 2347.
When low-temperature, thin films of either ionic or covalent
dinitrogen pentaoxide, N2O5, are exposed to gaseous HCl and
water, the only products observed in the solid phase by
reflection-absorption infrared spectroscopy (RAIRS) are mo-
lecular nitric acid and hydrated amorphous ionic nitric acid,
H3O+NO3-. Nitryl chloride, ClNO2, is not detectable. When
dinitrogen pentaoxide is co-deposited with hydrogen chloride
and water at 85 K, the resultant RAIR spectra indicate that the
film is composed of H3O+Cl-, N2O5, HNO3 and D2h-N2O4.
When nitryl chloride is co-deposited with either water or HCl/
water mixtures, infrared spectra indicative of solid D2h-N2O4
are measured, as well as peaks corresponding to nitrate ions
and cis-ClONO. Reaction between cis-ClONO and its isomer,
ClNO2, is proposed as an explanation for the formation of
dinitrogen tetraoxide in both systems. The proposed reaction
mechanism for this hydrolysis is extended to the N2O5/H2O/
HCl deposits in order to explain the lack of observable ClNO2
in such thin films.
Therefore, it appears that the low-temperature, solid-phase
chemistry of nitryl chloride isomers differs considerably from
room-temperature studies. These results serve to emphasize the
mechanistic importance of identifying reaction intermediates by
in situ surface measurements as a complement to experimental
monitoring of gas-phase reactants and products.
(34) Guillory, W. A.; Bernstein, M. L. J. Chem. Phys. 1975, 62, 1058.
(35) Ritzhaupt, G.; Devlin, J. P. J. Phys. Chem. 1991, 95, 90.
(36) Koch, T. G.; Horn, A. B.; Chesters, M. A.; McCoustra, M. R. S.;
Sodeau, J. R. J Phys. Chem. 1995, 99, 8362.
(37) Roddis, T. B. Ph.D. Thesis, University of East Anglia, 1998.
(38) St Louis, R. V.; Crawford, B. J. Chem. Phys. 1965, 42, 857.
(39) Fateley, W. G.; Bent, H. A.; Crawford, B. J. Chem. Phys. 1959,
31, 204.
(40) Durig, J. R; Kim, Y. H.; Guirgis, G. A. Spectrochim. Acta 1994,
50A, 463.
Acknowledgment. T.B.R. and M.P.G. thank EPSRC and
NERC (U.K.) for Ph.D. studentships, and J.R.S. thanks INTAS
for a grant (96-866).
(41) Tevault, D. E.; Smardzewski, R. R. J. Chem. Phys. 1977, 67, 3777.
(42) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. Chem.
Phys. Lett. 1978, 59, 78.
(43) Burrows, J. P.; Tyndall, G. S.; Moortgat, G. K. J. Phys. Chem.
1988, 92, 4340.
References and Notes
(1) Pandis, S. N.; Wexler, A. S.; Seinfeld, J. H. J. Phys. Chem. 1995,
99, 9646.
(2) Cadle, R. D.; Robbins, R. C. Discuss. Faraday Soc. 1960, 30, 155.
(3) Finlayson-Pitts, B. J. Nature 1983, 306 676.
(4) Finlayson-Pitts, B. J.; Johnson, S. N. Atmos. EnViron. 1988, 22,
1107.
(5) Livingston, F. E.; Finlayson-Pitts, B. J. Geophys. Res. Lett. 1991,
18, 17.
(44) Johnston, H. S.; Leighton, F. J. Am. Chem. Soc. 1953, 75, 3612.
(45) Chase, M. W.; Davies, C. A.; Downey, J. R.; Frurip, D. J.;
McDonald, R. A.; Syverud, A. N. J. Phys. Chem. Ref. Data 1985, 14 (Suppl.
1), 1.
(46) Horn, A. B.; Sodeau, J. R.; Roddis, T. B.; Williams, N. A. J. Phys.
Chem. A 1998, 102, 6107.