Gas-phase reactions of ClONO2 with Cl-(D2O)n = 0-3 and NO2-

Article abstract of DOI:10.1021/jp9701146

The reactions of ClONO₂ with the Cl^-(D₂O)_n≤3 and NO₂^- ions were studied in a flow-tube apparatus at several temperatures in the range 170-298 K and at a helium buffer gas pressure of 0.28 Torr. Rate constants for these reactions were determined, establishing that all these ions react quite efficiently with ClONO₂. The product ions observed in reaction sequences are reported, and reaction mechanisms are proposed to account for the findings. In the case of Cl^- and NO₂^- the main product ion is NO₃^-, while the hydrates Cl^-(D₂O)_n≤1-3 lead to production of the hydrated species, NO₃^-(D₂O)_m≤n-1. The main features of the observed secondary reactions can be described in terms of adduct formation and ligand switching. Limits on the bond energy of (NO₃ClNO₃)^-, 14.6 kcal/mol≤D(NO₃^--ClONO₂)≤26 kcal/mol, were determined. Finally, implications of these studies to atmospheric chemistry are discussed.

Full text of DOI:10.1021/jp9701146

8248
J. Phys. Chem. A 1997, 101, 8248-8254
Gas-Phase Reactions of ClONO2 with Cl-(D2O)n)0-3 and NO2
-
H. Wincel,† E. Mereand, and A. W. Castleman, Jr.*
Department of Chemistry, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802
ReceiVed: January 7, 1997; In Final Form: August 18, 1997X
The reactions of ClONO2 with the Cl-(D2O)ne3 and NO2 ions were studied in a flow-tube apparatus at
-
several temperatures in the range 170-298 K and at a helium buffer gas pressure of 0.28 Torr. Rate constants
for these reactions were determined, establishing that all these ions react quite efficiently with ClONO2. The
product ions observed in reaction sequences are reported, and reaction mechanisms are proposed to account
for the findings. In the case of Cl- and NO2- the main product ion is NO3 , while the hydrates Cl-(D2O)ne1-3
-
-
lead to production of the hydrated species, NO3 (D2O)men-1. The main features of the observed secondary
reactions can be described in terms of adduct formation and ligand switching. Limits on the bond energy of
(NO3ClNO3)-, 14.6 kcal/mol e D(NO3 -ClONO2) e 26 kcal/mol, were determined. Finally, implications
-
of these studies to atmospheric chemistry are discussed.
Introduction
Experimental Section
Both NO2 and ClONO2 are important components of the
reactive nitrogen reservoir; they play an important role in the
polar chemistry and are directly involved in the mechanism of
ozone destruction in the Antarctic stratosphere.1 Heterogeneous
reactions within or on the aerosol particles are thought to convert
the relatively unreactive reservoir species such as ClONO2 and
HCl into more active forms, Cl2 and HOCl, which are
photolyzed easily by sunlight to give Cl radicals that can
catalyze ozone destruction.2 In this respect, the reaction
All experiments were performed with a temperature variable
fast flow-tube (FT) apparatus, which has been described in detail
elsewhere;8b,c,9 hence, only a brief overview is presented here.
The reactant ion Cl- was formed by discharge ionization of
CCl4 in a flow-tube ion source using He as the buffer gas. The
-
NO2 ions were created when air (passed through liquid N2)
was added to the carrier gas. To produce the hydrates,
Cl-(D2O)n, a small amount of D2O was added to the CCl4/He
mixture. After formation, the ions were carried out by He
from the ion source into the FT, where they react with
preselected concentrations of ClONO2. The precursor and
product ions were mass analyzed by scanning the quadrupole
mass spectrometer, and ions were detected with a channeltron
electron multiplier. The rate constants were measured in the
usual way from the slope of the relative decrease in reactant
ion intensity with increasing ClONO2 flow rate, using ion
velocity values directly determined in pulsing experiments as
described earlier.9a Typical operating conditions encompassed
a total FT gas pressure of about 0.28 Torr, a buffer gas (He)
flow rate of 7 × 103 sccm, and FT temperatures ranging from
170 to 298 K. The FT temperature was controlled within an
accuracy of (1 °C.
ClONO2 + HCl f Cl2 + HNO3
(1)
is particularly important and its mechanism has evoked con-
siderable interest in various laboratories.3 Although reaction 1
is extremely slow in the gas phase4 (k < 10-19 cm3/s), it is
very efficient on water-ice and water-rich surfaces.3a-d It has
been suggested3a,e that ionic mechanisms could be involved in
the heterogeneous chemistry of the ClONO2/HCl system on
stratospheric aerosol particles. HCl molecules are expected to
exist at least partially dissociatively ionized, as Cl-(H2O)n in
an ice environment.3f,5 Thus, it is important to examine the
chemistry of the Cl-(H2O)n/ClONO2 ion/molecule reactions in
order to understand what effect these systems might have on
the mechanisms of ozone destruction. In a previous experi-
mental and theoretical study,6 the evidence for the rapid gas-
phase reaction of Cl- with ClONO2 has been presented, with a
rate constant k ) (9.2 ( 3) × 10-10 cm3/s at room temperature.
This value is much larger than that of k ) (9.6 ( 1) × 10-12
cm3/s, determined7 recently for the Cl/ClONO2 reaction of
neutral species. In the present work, the reactivities of the
Cl-(D2O)n)0-3 and NO2- ions toward ClONO2 at stratospheric
temperatures are investigated. We have previously reported8
the results of experimental studies on the cationic and anionic
water cluster reactions with N2O5, DNO3, Cl2O, and NO. A
motivating factor in these studies is, in part, an interest in the
possible ionic mechanisms in heterogeneous reactions that may
be of importance in stratospheric chemistry. To the best of our
knowledge, there has been no prior report of the reactivity of
hydrated Cl- ions with ClONO2.
The ClONO2 was synthesized by the reaction of Cl2O with
N2O5 on the basis of the method of Schmeisser10 and then
purified. The Cl2O and N2O5 were synthesized and purified
by procedures described previously.8b,e Care was taken to
eliminate all moisture from the synthesis apparatus and the
experimental gas inlet in order to minimize the heterogeneous
reactions: N2O5 + H2O f 2HNO3, Cl2O + H2O f 2HOCl,
and ClONO2 + H2O f HNO3 + HOCl. The ClONO2 samples
were stored in the dark at liquid nitrogen temperature to prevent
decomposition. The purity of the ClONO2 samples was
+
analyzed using the charge-transfer reaction(s) of O2 with
ClONO2 + impurities (Cl2, Cl2O, HNO3). The ionization
potential11 of O2 (12.071 eV) is higher than those of Cl2 (11.48
+
eV), Cl2O (10.94 eV), and HNO3 (11.95 eV). No Cl2 and
Cl2O+ signals have been found in the mass spectra (Figure 1)
of our purified ClONO2 samples in the FT. Therefore, we
believe the impurity levels of Cl2 and Cl2O in ClONO2 to be
<0.2%. We attribute the major ion NO2+, together with a small
amount of NO+ detected in the spectra (Figure 1), to the
dissociative charge-transfer reactions
† Permanent address: Institute of Physical Chemistry, Polish Academy
of Sciences, Kasprzaka 44/52, Warsaw, Poland.
X Abstract published in AdVance ACS Abstracts, October 1, 1997.
S1089-5639(97)00114-X CCC: $14.00 © 1997 American Chemical Society

Reactions of ClONO2
J. Phys. Chem. A, Vol. 101, No. 44, 1997 8249
Figure 1. Mass spectra registered before (inset) and after addition
through the Teflon line of 70 sccm of the ClONO2/He mixture into the
flow tube by eluting the ClONO2 vapor with a stream of He from the
cooled trap at 156 K (procedure 1; see Experimental Section). Flow
tube temperature is 180 K, and flow tube pressure is 0.28 Torr.
O2+ + ClONO2 f NO2+ + ClO + 25.6 kcal/mol
f NO+ + ClO2 + 25 kcal/mol
(2a)
(2b)
It is worth mentioning that we also examined the ClONO2
sample prepared by the reaction of Cl2O with excess N2O5, and
the results of the experiments were similar to those for the
sample presented in Figure 1. These experiments indicate that
a potential N2O5 impurity in our purified ClONO2 samples was
Figure 2. Observed variation of ion signals for the addition through
the Teflon line of the 0.47% ClONO2/He mixture into the flow tube
(procedure 2). T ) 170 K, P ) 0.28 Torr.
stream of dry He (hereafter referred to as procedure 1); (2)
introduction as the ClONO2/He mixture (typically 0.3-0.5%
ClONO2/He) prepared in a 1 L glass reservoir and during the
experiments stored at 273 K using an ice bath (referred to as
procedure 2). In both cases the flow rate of the mixture was
controlled with an MKS flow meter; the flow rate was calibrated
using the ClONO2 heat capacity reported by Miller et al.12
Chemicals. He (Airco, 99.5%), Cl2 (Matheson, 99.99%), and
O2 (Linde, 99.9%) were used without further purification.
+
largely removed. The fact that no ClO+ or ClO2 signals are
seen in the mass spectrum of O2+/(ClONO2 + impurities)
suggests that the decomposition of ClONO2 into ClO + NO2
and ClO2 + NO is negligible under the present experiments.
+
The small ion signals at m/e 63 and 143 are attributed to HNO3
(from impurity) and NO2+ClONO2 (from the NO2+/ClONO2
reaction), respectively. The concentrations of XNO3 (X ) H
and D) in ClONO2 in the FT were determined via the
-
termolecular reaction NO3 + XNO3 + He using the rate
Results and Discussion
constant value from ref 8f. The relative concentrations of XNO3
and ClONO2 were also estimated by measuring the ratios of
the product ions [(NO3XNO3)-]/[(NO3ClNO3)-] from reactions
6 and 7, using the rate constant ratio (k6/k7) ) 0.055 on the
basis of the reported8f,17 values and assuming the same tem-
perature dependence. To minimize the interference of the
switching reaction 14a, in both cases, the measurements were
performed at low flow rates of the ClONO2/He mixture. In
the experiments described in this paper, at low flow rates the
ratio [(NO3XNO3)-]/[(NO3ClNO3)-] varied between 0.02 and
0.1 and the level of impurity of XNO3 in ClONO2 in the FT
was <0.6%. However, when a small amount of D2O was
introduced into the ion source to produce Cl-(D2O)n, the
formation of a significant fraction of (NO3DNO3)- and
NO3-DOCl, along with that of (NO3ClNO3)-, was observed.
There are various possible sources of the (NO3DNO3)- ion, one
being that ClONO2 may react with D2O with surfaces of the
FT to give DNO3 and DOCl, the products of which may in
turn become clustered to NO3-. We estimate that as high as
15% of ClONO2 could be hydrolyzed in the FT owing to the
addition of D2O to the ion source. For this reason we used the
smallest concentration of D2O possible in the ion source for
the creation of the Cl-(D2O)n)1-3 ion clusters and for the
measurements of their reactivities. ClONO2 was introduced into
the FT through a Teflon, as well as via a stainless steel, reactant
gas inlet (RGI) line. Two procedures for introducing this
reactant were utilized: (1) ClONO2 vapor was eluted from a
cooled trap at 156 K (ethanol/liquid N2 cooling bath) with a
Reactions of Cl-(D2O)n)0-3. The observed reactivity of Cl-
with ClONO2 is illustrated by the data shown in Figure 2. The
-
primary ion from the Cl-/ClONO2 reaction is NO3 (Figures
2-4). This observation is in agreement with the earlier study6
showing that this product ion is formed by reaction 3.
(3)
The presence of HNO3 could lead to the formation of a low-
intensity ion complex (NO3HNO3)-, which is observed along
with the major product (NO3ClNO3)- (Figures 2-4). In the
-
absence of hydrated NO3 complexes, the latter product ion
must arise via a three-body association reaction under the
conditions of these experiments. However, it should be noted
-
that HNO3 can also lead to the NO3 ion product in the Cl-/
HNO3 primary reaction (k ) 1.6 × 10-9 cm3/s).16 Ab initio
calculations6 on the Cl-/ClONO2 system show that reaction 3
is initiated by the interaction of Cl- with the positively charged
Cl atom of ClONO2 followed by the formation of the intermedi-
ate ion-molecule complex, Cl2‚NO3-, which dissociates to the
-
NO3 + Cl2 products.

8250 J. Phys. Chem. A, Vol. 101, No. 44, 1997
Wincel et al.
Figure 3. Mass spectra obtained before (inset) and after addition
through the stainless steel line (procedure 2) of 47 sccm of the ClONO2/
He mixture into the flow tube at T ) 200 K and a flow tube pressure
of 0.28 Torr.
Figure 5. Mass spectra obtained before (a) and after (b) addition
through the stainless steel line (procedure 2) of 20 sccm of the ClONO2/
He mixture into the flow tube at T ) 170 K and a flow tube pressure
of 0.28 Torr.
Figure 4. Mass spectra obtained before (inset) and after addition
through the Teflon line (procedure 2) of 8 sccm (a) and 78 sccm (b) of
the 0.49% ClONO2/He mixture into the flow tube at T ) 170 K and a
flow tube pressure of 0.28 Torr.
The data in Figures 5 and 6 indicate that when the
Cl-(D2O)n)1-3 ions are allowed to react with ClONO2, the
hydrated species NO3-(D2O)ne2 are formed as the primary
product ions. These data suggest that the following reactions
occur under the present experimental conditions
Figure 6. Observed variation of ion signals for the addition through
the stainless steel line (procedure 2) of the ClONO2/He mixture into
the flow tube at T ) 200 K and P ) 0.28 Torr.
Cl-(D2O)n + ClONO2 f [(D2O)nCl-‚‚‚ClONO2]* f
-
NO3 (D2O)m + Cl2 + (n - m)D2O (4)
As can be seen from an examination of Table 1, reaction 4
is significantly exothermic for (n - m) e 2. However, the
excited ions NO3-(D2O)n when formed from these reactions
have sufficient energy to dissociate in the FT by losing only
The neutral products are not observed in the present experiments
but are only inferred from the mass balance and thermochemical
information.11

Reactions of ClONO2
J. Phys. Chem. A, Vol. 101, No. 44, 1997 8251
TABLE 1: Enthalpy Changes, ∆H (kcal/mol),a for the
Reaction Cl-‚(D2O)n + ClONO2 f NO3-‚(D2O)m + Cl2 + (n
- m)D2O
one D2O molecule. In the case of NO3-(D2O)3, simple RRK
calculation14 indicates that the process of dissociation by loss
of two D2O molecules is much longer than the average ion
residence time in the FT.
m
It should be noted that any nitric acid present in the system
(as impurity or from hydrolysis of ClONO2) can undergo
exothermic reaction 5, also leading to the NO3-(D2O)n product
ions:
n
0
1
2
3
0
1
2
3
-25.5
-10.6
2.3
-25.2
-12.3
-0.6
-26.6
-14.9
14.0
-28.7
-
Cl-(D2O)n + XNO3 f NO3 (D2O)n + XCl
(5)
a All ∆H values were estimated on the basis of the thermodynamic
data cited in ref 11.
where X ) D or H.
Reactions of NO3-(D2O)n)0-2. From the data shown in
molecule(s) draw(s) more negative charge than other oxygen-
(s) of NO3-. Hence, it might be expected that such oxygen
atom(s) would be the most likely site(s) for the initial attack on
the positively charged Cl atom of ClONO2 to form the
intermediate similar to that in reaction 8. It is likely that the
analogous intermediate structure is also involved in the well-
known displacement reaction 9, which is evident from the data
in Figure 6.
-
Figures 2 and 4 we deduce that the primary product ion NO3
associatively reacts with both ClONO2 and HNO3 by adduct
formation as indicated by reactions 6 and 7, respectively.
NO3- + ClONO2 T [(NO3ClNO3)-]*
9
He8
(NO3ClNO3)- (6)
NO3- + HNO3 T [(NO3HNO3)-]*
9
He8 (NO3HNO3)- (7)
-
NO3 D2O + DNO3 f (NO3DNO3)- + D2O
(9)
Considering the origin of the species (NO3HNO3)-, reaction 7
is one possibility. (Other possibilities could be reactions 9, 10a,
11b, and 14b discussed below.) On the other hand, our studies
indicate that very little impurity HNO3 is present and that the
surface hydrolysis of ClONO2 is expected to be slow under the
present experimental conditions. Furthermore, any produced
HNO3 would be expected to remain largely adsorbed on the
walls of the flow tubes. Reaction 6 has been observed
previously, and the rate constant values at 232 K in H2 (k ) 3
× 10-10cm3/s)16a and at 298 K in He (k ) 3 × 10-11cm3/s)17
have been reported. It has been reported elsewhere15 that
reaction 7 is termolecular under FT conditions similar to those
in the present study, and we anticipate that this is also in the
case of reaction 6.
We cannot rule out the reaction
-
NO3 D2O + ClONO2 f (NO3DNO3)- + DOCl
(10a)
(10b)
-
f NO3 DOCl + DNO3
Although SIFT experiments by van Doren et al.16b did not show
the occurrence of reaction 10, Schindler et al.16c did observe
this reaction utilizing ICR. Recent ab initio calculations19 by
Doyle and Dunlap indicate that the C2h isomer shown below,
which has its C2 axis perpendicular to the plane of molecule, is
the lowest energy structure of the (NO3HNO3)- system.
When the hydrated ion NO3-(D2O) is produced as a result
of the Cl-(D2O)n/ClONO2 reactions discussed above, another
possible pathway, leading to the formation of (NO3ClNO3)- and
(NO3DNO3)- complexes, is D2O displacement from NO3-(D2O)
by ClONO2 and DNO3, respectively. Such reactions previously
have been observed in other laboratories.13,16 It should be
mentioned that in the case of the Cl-(D2O)n/ClONO2 study, the
formation of the NO3-(DOCl) complex is also observed, but at
200 K, its intensity is much lower than that of (NO3DNO3)-
(see Figure 5b). However, at temperatures about 170 K the
intensities of NO3-(DOCl) and (NO3DNO3)- are comparable.
A possible mechanism for the NO3-D2O/ClONO2 reaction
is as follows.
This isomer is almost isoenergetic with an isomer having
a C2 structure, which has its C2 axis in the plane of the
molecule.
A related possibility (also not seen by Van Doren et al.16b
)
is the reaction of the product ion (NO3ClONO2)- with
water vapor, which might occur, since this ion is in large
abundance:
-
(NO3ClONO2)- + D2O f NO3 DOCl + DNO3
(11a)
f (NO3DNO3)- + DOCl (11b)
The data in Figure 6 shows that the total intensities of the
NO3-, (NO3ClNO3)-, and (NO3DNO3)- product ions exceed
the intensities of the precursor ions Cl- and Cl-(D2O).
Likewise, the data in Figure 2 show that the product ion
intensities may slightly exceed those of the reactants. This is
most likely due to the dissociative electron attachment process:
-
eth- + ClONO2 f [ClONO2 ]* f products
(12)
At moderate concentration of a neutral reactant (6 × 1010
-
cm-3), the reaction leads20 primarily to the NO2- (∼50%), NO3
Ab initio calculations18 on NO3-(H2O)n)1-3 indicate that the
(∼30%), and ClO- (∼20%) product ions, together with minor
ions ClONO2- (e2%) and Cl- (e6%) whose origin is not clear.
oxygen atom(s) of NO3- forming hydrogen bond(s) with water

8252 J. Phys. Chem. A, Vol. 101, No. 44, 1997
Wincel et al.
as those shown in reactions 146 and 15.
Figure 7. Observed variation of the ratio [(NO3DNO3)-]/[(NO3-
ClNO3)-] with flow rate of the ClONO2/He mixture introduced into
the flow tube through the stainless steel line (procedure 2) at T ) 200K
and P ) 0.28 Torr.
(NO3DNO3)- + ClONO2 T (NO3DNO3)-ClONO2 (15)
Both reactions 14b and 15 require collisional stabilization. It
should be mentioned that the reaction 14a has been observed16a,b
directly with a selected ion flow drift tube apparatus at 283 K.
The observation of switching reactions 8 and 14a places limits
on the bond energy of (NO3ClNO3)-: 14.6 kcal/mol <
D(NO3--ClONO2) < 26 kcal/mol.23 This is consistent with
an energy of 21 kcal/mol calculated24 at the higher MP2
6-31+G(d) level of theory. Formation of the (NO3DNO3)-DNO3
ion complex, shown in Figure 6, is possibly attributed to the
well-known8f association reaction 16
-
At higher ClONO2 concentration (6 × 1011 cm-3) the NO2
,
-
ClO-, and Cl- ions quickly react with ClONO2, leading to NO3
-
and clusters of NO3 with ClONO2 and the impurity HNO3.20
The rate constant values (k ) 3.9 × 10-9 cm3/s,21 k ) 1.1 ×
10-7 cm3/s 20) reported for electron attachment to ClONO2
suggest that such a process could be important in the present
case if the density of free electrons in the afterglow is relatively
high. In contrast to the NO2-/Cl2O study,8e in the present
measurements our attempts to eliminate the observed effect
(Figure 6 and, to a much lesser extent, Figure 2) by minimization
of the discharge ionization in a flow-tube ion source were not
very successful presumably because of the much higher ef-
ficiency of electron attachment to ClONO2 compared with Cl2O
(k ) 2.6 × 10-10 cm3/s).22 Any NO3- ions produced by direct
dissociative electron attachment process (reaction 12) or by
secondary ion chemistry, similar to those arising from reaction
3, can undergo secondary reactions with ClONO2 and XNO3
(impurity) to yield the (NO3ClNO3)- and (NO3XNO3)- product
ions.
(NO3DNO3)- + DNO3 T (NO3DNO3)-DNO3 (16)
However, we cannot exclude that such a complex also can be
formed in the following ligand-switching reaction
(NO3DNO3)-ClONO2 + DNO3 T
(NO3DNO3)-DNO3 + ClONO2 (17)
or via the reaction
The general behavior of the NO3-(D2O)2 ion cluster and the
observed formation of (NO3ClONO2)-D2O) (Figure 6) suggest
that reaction 13 occurs as follows:
(NO3DNO3)-ClONO2 + D2O f
(NO3DNO3)-DNO3 + DOCl (18)
which does not require collisional stabilization.
Reactions of NO2-. Our data (Figure 2) clearly indicate that
-
NO2 reacts efficiently with ClONO2, and the major product
-
ion from this, as well as from that of Cl-/ ClONO2, is NO3
.
From careful consideration of the data, we conclude that the
principle reaction channel of the NO2-/ClONO2 system is
NO2- + ClONO2 f NO3- + ClONO
(19a)
This reaction is exothermic by 14.8 kcal/mol (reaction 11a). It
-
is worthwhile to point out that when NO2 is present as the
-
reactant ion, a small amount of the ClONO2 product ion is
also observed (Figure 4a). This could be attributed to the charge
transfer from NO2 to ClONO2 by reaction channel 19b:
-
Observation of this reaction suggests that the bond energy
D(NO3-D2O-ClONO2) > D(NO3-D2O-D2O) ) 14.3 kcal/
mol.11b Note that the intensity of this ion is very small in accord
with its expected reactive nature toward disproportionation in
analogy to reactions 11a and 11b.
Reactions of (NO3ClNO3)- and (NO3DNO3)-. As can be
seen in Figure 7, the observed ionic product ratio [(NO3-
DNO3)-]/[(NO3ClNO3)-] increases with the flow rate of the
(ClONO2/He) mixture. This could be attributed to reaction 11b,
through this behavior, together with our observation of the ion
at m/e 223, which could be (NO3ClNO3)-DNO3 and/or
(NO3DNO3)-ClONO2 possibly due to reaction mechanisms such
NO2- + ClONO2 f ClONO2- + NO2
(19b)
It should be mentioned that both reaction channels 19a and 19b
have been observed previously by Van Doren et al.17 and the
rate constant for these channels reported as k ) (1.5 ( 40%) ×
10-9 cm3/s. As can be seen in Table 2, this value is significantly
larger than ours but remains within the range of experimental
uncertainty. Thermochemical information indicates that dis-
sociation of ClONO2 from channel 19b into NO3- + Cl or Cl-
+ NO3 is not energetically accessible. The observation of

Reactions of ClONO2
J. Phys. Chem. A, Vol. 101, No. 44, 1997 8253
TABLE 2: Rate Constantsa for the Reactions of ClONO2
measurements, at low flow rates (<20 sccm) the decrease of
Cl- with the ClONO2 concentration was observed to be lower
than that at higher flow rates where the decrease is linear. Such
behavior may suggest that Cl- could be produced in the FT by
interaction with free electrons. For the experiments described
here, this effect was minimized (Figures 2 and 6) by optimizing
the discharge in the ion source. Therefore, we believe that
channel 12 is not a significant contributor to the errors in the
measured values of kexp for Cl-/ClONO2. As seen in Table 2,
kexp for this reaction is very close to the value of Okumura et
al.6 determined when ClONO2 was introduced into the ICR
apparatus through the glass inlet system. From the table it is
evident that kexp values are close to kcal and that the kexp/kcal
ratio decreases slightly with hydration. These data seem to
suggest that energy barriers exist in the potential surfaces for
-
with Cl-(D2O)n)0-3 and NO2
kexp
reactant
b
ion
this work
reported
kcal
kexp/kcal T (K)
Cl-
1.16 ( 0.3 (0.74 ( 0.3)c,f
(1.13 ( 0.37)d,f
(0.92 ( 0.3)e,f
1.04 ( 0.3
1.39
0.83
298
1.48
1.53
1.27
1.31
1.16
1.19
1.09
1.12
1.34
1.39
0.70
0.72
0.82
0.61
0.73
0.61
0.64
0.54
0.75
0.70
200
170
200
170
200
170
200
170
200
170
1.10 ( 0.2
Cl-(D2O)
1.04 ( 0.3
0.80 ( 0.2
Cl-(D2O)2 0.85 ( 0.3
0.73 ( 0.2
Cl-(D2O)3 0.70 ( 0.3
0.61 ( 0.2
-
NO2
1.01 ( 0.3 (1.5 ( 0.6)+2
0.98 ( 0.2
the Cl-(D2O)n)0-3/ClONO2 reactions. Presumably, the kexp
/
a Units for all rate constants are 10-9 cm3/s. b Rate constants were
calculated from the Su-Chesnavich theory,25 using a ) 6.28 Å3 and
µD ) 0.77 D. c Rate constant measured using the stainless steel gas
inlet system. d Rate constant measured using the glass inlet system.
e Average value from the data obtained with stainless steel and glass
inlet systems. f Rate constant measured at 298 K.
kcal trend is a reflection of the solvent reorganization about
the new charge distribution in the intermediate complex
[(D2O)nCl-ClONO2] to the observed products NO3-(D2O)m +
neutrals. Similarly, the (kexp/kcal) ) 0.75 ratio for NO2-/ClONO2
implies that some barrier exists for the forward reaction channel.
This situation is similar to that found for the NO2-/N2O5 system,
where the reaction efficiency is kexp/kcal ) 0.68 at 231 K. These
observations may reflect significant intramolecular rearrange-
ments within the intermediate complexes of both reactions.
charge transfer in reaction 19b places a limit on the electron
affinity (EA) of ClONO2; EA(ClONO2) is greater than that of
NO2-. This is in agreement with the findings of van Doren et
al.17
Atmospheric Implications. As already mentioned, the
reaction of ClONO2 with HCl is an important heterogeneous
reaction on polar stratospheric clouds (PSCs), which is thought
to convert the Cl trapped in both these species into an easily
photolizable form, Cl2. HCl has been proposed to exist on
PSCs, at least partially, as Cl-(H2O)n.3f The results presented
here indicate that Cl-(D2O)ne3 reacts quite efficiently with
ClONO2, leading to Cl2 and the stable ions NO3-(DNO3)n
by a series of sequential reactions. The NO3-(HNO3)n)1,2
represent the dominant anions found27 through much of the
stratosphere and troposphere. Thus, it would appear that an
ionic mechanism can be involved in the heterogeneous pro-
cess, leading to conversion of ClONO2/HCl into Cl2 onto
PSCs. However, implying the possible importance of this
mechanism in stratospheric chemistry requires caution, as well
as further considerations. First, the Cl-(H2O)n/ClONO2 reaction
efficiency decreases with increasing cluster size. The distribu-
tions of both the Cl-(H2O)n ion clusters and ClONO2 withinsor
on the surface ofswater droplets are difficult to predict.
Second, we do not contend that the observed gas-phase
reactivities of hydrated Cl- are a direct measure of their
reactivity in the condensed state of PSCs. Third, the reactivities
of Cl-(H2O)n toward other important neutrals (e.g., HOCl and
HNO3) that are expected to be present on PSCs have not been
assessed.
Rate Constants. The rate constants for the studied systems
are summarized in Table 2. The table also includes the values
reported by Okumura et al.6 for Cl-/ClONO2 at room temper-
ature as well as the calculated ones from the Su-Chesnavich
theory25 using the calculated polarized ability24 of 6.28 × 10-24
cm3 and dipole moment26 of 0.77 D. In the present experiments
the kexp values were obtained using the Teflon inlet line (at T
) 170 and 298 K), the stainless steel line (at T ) 200 K), and
procedure 2. In every case these values are an average of at
least four measurements, and the error limits show the statistical
fluctuations. For the Cl-/ClONO2 and NO2-/ClONO2 systems,
an absolute uncertainty in the kexp values can be as large as
30%, while that for Cl-(D2O)n ) 1-3/ClONO2 may be as high
as 60%. The major factors that influence the accuracy of the
measured rate constants are (1) contributions of impurities from
the synthesis and/or formed in the FT and (2) interferences of
products from the reactions of free electrons with ClONO2 in
the FT that may influence the measured slope of the ion decay
plots vs ClONO2 flow rate.
In the rate constant measurements of Cl-/ClONO2 and NO2
-
/
ClONO2, the amount of HNO3 impurity in ClONO2 introduced
into the FT was estimated to be <0.6%. The presence of such
a level of HNO3 would lead to an enhancement of less than
0.4% in kexp due to the Cl-/HNO3 reactions (k ) 1.6 × 10-9
cm3/s 13) and NO2-/HNO3 (k ) 1.6 × 10- 9cm3/s 13). However,
as discussed in detail, for the Cl-(D2O)n)1-3/ClONO2 cases,
when a small amount of D2O was introduced into the ion source
to produce the hydrated reactant ions Cl-(D2O)n)1-3, the
concentration of nitric acid (DNO3) might significantly increase
owing to the hydrolysis of ClONO2 in the FT. This could lead
to observation of higher kexp values (up to 10%). Since we did
not find any evidence for the Cl-(D2O)n)0-3/DOCl reactions
in the present experiments, it was assumed that the DOCl formed
from the hydrolysis of ClONO2 does not affect our measured
Conclusions
-
The reactivity of the Cl-(D2O)ne3 and NO2 ions with
ClONO2 has been investigated. The efficiency of the Cl--
(D2O)ne3/ClONO2 reactions was observed to decrease with
increasing cluster size. Sequential chemistry initiated by these
reactions has been presented. These reactions provide pathways
for the conversion of Cl from Cl-(H2O)n/ClONO2 to photo-
chemically labile Cl2 and mechanisms for the formation of the
NO3-(HNO3)n ionic complexes. The NO2-/ClONO2 reaction
was found to lead predominantly to the NO3- ion product, but
a small fraction of charge transfer in this reaction was also
observed.
rate constants for Cl-(D2O)n)1-3/ClONO2. In the case of Cl-/
-
ClONO2 and NO2-/ClONO2, the interference of Cl- and NO2
,
respectively, from reactions with free electrons cannot be
excluded. In the present study, we varied ion-source conditions
in order to obtain evidence for such a possibility. In some

8254 J. Phys. Chem. A, Vol. 101, No. 44, 1997
Wincel et al.
(9) Upschulte, B. L.; Schul, R. J.; Passarella, R.; Keesee, R. G.;
Castleman, A. W., Jr. Int. J. Mass Spectrom. Ion Processes 1987, 75, 27.
(b) Yang, X.; Zhang, X.; Castleman, A. W., Jr. Int. J. Mass Spectrom. Ion
Processes 1991, 109, 339.
Acknowledgment. Financial support by the Atmospheric
Sciences Division of the National Science Foundation, Grant
No. ATM-9321660, is grateful acknowledged. We are also
grateful to the National Science Foundation for a grant through
the International Program Office, Grant No. INT-9208271.
(10) Schmeisser, M. Inorg. Synth. 1967, 9, 127.
(11) Unless stated otherwise, all thermochemical information has been
taken or derived from the following. (a) Lias, S. G.; Bartness, J. E.; Liebman,
J. F.; Holmes, J. L.; Levin, R. D.; Mallard, W. G. J. Phys. Chem. Ref.
Data, Suppl. 1 1988, 17. (b) Keesee, R. G.; Castleman, A. W., Jr. J. Phys.
Chem. Ref. Data 1986, 15, 1011. (c) Hiraoka, K.; Mizuse, S.; Yamabe, S.
J. Phys. Chem. 1988, 92, 3943.
(12) Miller, R. H.; Bernitt, D. L.; Hisatsune, I. C. Spectrochim. Acta
1967, 23A, 223.
(13) Fehsenfeld, F. C.; Howard, C. J.; Schmeltekopf, A. L. J. Chem.
Phys. 1975, 63, 2835.
(14) The calculations employ classical RRK theory (Chemical Kinetics
and Dynamics; Steinfeld, J. L., Francisco, J. D., Hase, W. L., Eds.; Prentice-
Hall: Englewood Cliffs, NJ, 1989; p 358) for the prediction of the lifetime,
t ) 1/k, against unimolecular decomposition of the excited NO3-‚(D2O)n,
where the value of the internal energy, E, of these species was taken as
that estimated from the distribution of the exothermicity between ionic,
NO3-‚(D2O)n, and neutral products of reaction 4. The critical energy for
the decomposition process, E°, was taken as being equal to the bond
dissociation energy D[NO3-‚(D2O)n-1-D2O].
(15) (a) Davidson, J. A.; Fehsenfeld, F. C.; Howard, C. J. Int. J. Chem.
Kinet. 1977, 9, 17. (b) Viggiano, A. A.; Dale, F.; Paulson, J. F. J. Geophys.
Res. 1985, 90, 7977.
(16) (a) Viggiano, A. A.; Morris, R. A.; Van Doren, J. M. J. Geophys.
Res. 1994, 99, 8221. (b) Van Doren, J. M.; Viggiano, A. A.; Morris, R. A.
J. Am. Chem. Soc. 1994, 116, 6957. (c) Schindler, T.; Berg, C.; Niedner-
Schatteburg, G.; Bondybey, V. E. J. Chem. Phys. 1996, 104, 3998.
(17) Van Doren, J. M.; Viggiano, A. A.; Morris, R. A.; Miller, T. M. J.
Chem. Phys. 1995, 103, 10806.
References and Notes
(1) (a) Solomon, S. Nature 1990, 347, 347. (b) Brune, W. H.; Anderson,
J. G.; Toohey, D. W.; Fahey, D. W.; Kawa, S. R.; Jones, R. L.; McKenna,
D. S.; Poole, L. R. Science 1991, 252, 1260.
(2) (a) Webster, C. R.; May, R. D.; Toohey, D. W.; Avallone, L. M.;
Anderson, J. G.; Newman, P.; Lait, L.; Schoeberl, M. R.; Elkins, J. W.;
Chan, K. R. Science 1993, 261, 1130. (b) Toohey, D. W.; Avallone, L. M.;
Lait, L. R.; Newman, P. A.; Schoeberl, M. R.; Fahey, D. W.; Woodbridge,
E. L.; Anderson, J. G. Science 1993, 261, 1134. (c) Toon, O.; Browell, E.;
Gary, B.; Lait, L.; Livingston, J.; Newman, P.; Pueschell, R.; Russell, P.;
Schoeberl, M.; Toon, G.; Traub, W.; Valero, F. P. J.; Sekirk, H.; Jordan, J.
Science 1993, 261, 1136.
(3) (a) Molina, M. J.; Tso, T.-L.; Molina, L. T.; Wang, F. C.-Y. Science
1987, 238, 1253. (b) Tolbert, M. A.; Rossi, M. J.; Malhorta, R.; Golden,
D. M. Science 1987, 238, 1258. (c) Hanson, D. R.; Ravishankara, A. R. J.
Geophys. Res. 1991, 96, 5081. (d) Hanson, D. R.; Ravishankara, A. R. J.
Phys. Chem. 1992, 96, 2682. (e) Sodeau, J. R.; Horn, A. B.; Banham, S.
F.; Koch, T. G. Science 1995, 99, 6258. (f) Gertner, B. J.; Hynes, J. T.
Science 1996, 271, 1563.
(4) Molina, L. T.; Molina, M. J.; Stachnik, R. A.; Tom, R. D. J. Phys.
Chem. 1985, 89, 3779.
(5) (a) Yoon, Y. K.; Carpenter, G. B. Acta Crystallogr. 1959, 12, 17.
(b) Lundgren, J. O.; Olorsson, I. Acta Crystallogr. 1967, 23, 966. (c) Taesler,
I.; Lundgren, J. O. Acta Crystallogr. 1978, 34, 2424.
(18) Howell, J. M.; Sapse, A. M.; Singman, E.; Snyder, G. J. Chem.
Phys. 1982, 86, 2345.
(6) Hass, B.-M.; Crellin, K. C.; Kuwata, K. T.; Okumura, M. J. Phys.
Chem. 1994, 98, 6740.
(19) Doyle, R. J.; Dunlap, B. I. J. Phys. Chem. 1994, 98, 8261.
(20) Van Doren, J. M.; McClellan, J.; Miller, T. M.; Paulson, T. F.;
Viggiano, A. A. J. Chem. Phys. 1966, 105, 104.
(21) Wecker, D.; Schindler, R. N. Z Naturforsch., A 1984, 39, 342.
(22) Wecker, D.; Christodoulides, A. A.; Schindler, R. N. Int. J. Mass
Spectrom. Ion Phys. 1981, 38, 391.
(7) Yokelson, R. J.; Burkholder, J. B.; Goldfarb, L.; Fox, R. W.; Gilles,
M. K.; Ravishankara, A. R. J. Phys. Chem. 1995, 99, 13976.
(8) (a) Zhang, X.; Mereand, E.; Castleman, A. W., Jr. J. Phys. Chem.
1994, 98, 3554. (b) Wincel, H.; Mereand, E.; Castleman, A. W., Jr. J. Phys.
Chem. 1994, 98, 8606. (c) Wincel, H.; Mereand, E.; Castleman, A. W., Jr.
J. Phys. Chem. 1995, 99, 1792. (d) Wincel, H.; Mereand, E.; Castleman,
A. W., Jr. J. Chem. Phys. 1995, 102, 9228. (e) Wincel, H.; Mereand, E.;
Castleman, A. W., Jr. J. Phys. Chem. 1995, 99, 15678. (f) Wincel, H.;
Mereand, E.; Castleman, A. W., Jr. J. Phys. Chem. 1996, 100, 7488. (g)
Wincel, H.; Mereand, E.; Castleman, A. W., Jr. J. Phys. Chem. 1996, 100,
16808.
(23) The bond dissociation energy of NO3-HNO3 is 26 kcal/mol. The
value for NO3-H2O is 14.6 kcal/mol. The relevant data are found in ref 11.
(24) Mebel, A. M.; Morokuma, K. J. Phys. Chem. 1996, 100, 2985.
(25) Su, T.; Chesnavich, W. J. J. Chem. Phys. 1982, 76, 5183.
(26) Suenram, R. D.; Johnson, D. R. J. Mol. Spectrosc. 1977, 65, 239.
(27) (a) Haitman, H.; Arnold, F. Nature 1983, 306, 747. (b) Krieger,
A.; Arnold, F. Geophys. Res. Lett. 1994, 21, 1259.