HONO solubility and heterogeneous reactivity on sulfuric acid surfaces

Article abstract of DOI:10.1021/jp9807120

The solubility of HONO in various wt% sulfuric acid solutions was measured. HONO becomes increasingly more soluble in high acid wt% solutions, probably resulting from the protonation of HONO to give H₂ONO⁺, followed by dissociation and association to give NO⁺HSO₄^- (nitrosyl sulfuric acid). The reaction of HONO with HCl on 50, 60, 67, and 70 wt% sulfuric acid solutions was also studied. The measured uptake coefficients could not be explained simply by the occurrence of a bulk phase reaction, and a surface effect is found to contribute to the uptake. The reaction of HONO with HCl was measured on sulfuric acid particles, and a small reacto-diffusive length further supports a surface reaction. From extrapolating our measured uptake values to stratospheric conditions, the formation of ClNO from the heterogeneous reaction of HONO and HCl is concluded to be an insignificant process in the volcanically unperturbed stratosphere.

Full text of DOI:10.1021/jp9807120

J. Phys. Chem. A 1998, 102, 3323-3332
3323
HONO Solubility and Heterogeneous Reactivity on Sulfuric Acid Surfaces
C. A. Longfellow, T. Imamura,† A. R. Ravishankara,‡ and D. R. Hanson*
National Oceanic and Atmospheric Administration, Aeronomy Laboratory, 325 Broadway,
Boulder, Colorado 80303, and CooperatiVe Institute for Research in EnVironmental Sciences,
UniVersity of Colorado, Boulder, Colorado
ReceiVed: January 7, 1998; In Final Form: March 6, 1998
The solubility of HONO in various wt % sulfuric acid solutions was measured. HONO becomes increasingly
more soluble in high acid wt % solutions, probably resulting from the protonation of HONO to give
H2ONO+, followed by dissociation and association to give NO+HSO4- (nitrosyl sulfuric acid). The reaction
of HONO with HCl on 50, 60, 67, and 70 wt % sulfuric acid solutions was also studied. The measured
uptake coefficients could not be explained simply by the occurrence of a bulk phase reaction, and a surface
effect is found to contribute to the uptake. The reaction of HONO with HCl was measured on sulfuric acid
particles, and a small reacto-diffusive length further supports a surface reaction. From extrapolating our
measured uptake values to stratospheric conditions, the formation of ClNO from the heterogeneous reaction
of HONO and HCl is concluded to be an insignificant process in the volcanically unperturbed stratosphere.
Introduction
reaction could be important in the stratosphere. They found a
reaction probability ranging from 0.01 to 0.02 for HCl uptake
Heterogeneous activation of chlorine by polar stratospheric
clouds is a well-documented phenomena1,2 and has been
extensively studied by field campaigns3 and in the laboratory.4
In the extrapolar mid to lower stratosphere (15-25 km), sulfate
aerosols of 60-80 wt % sulfuric acid are the predominant
heterogeneous surfaces5 and the heterogeneous activation of
chlorine is thought to be much less efficient.6 For example,
reactions such as ClONO2 + HCl and HOCl + HCl can occur
in sulfuric acid, but their efficiencies decrease considerably as
temperatures warm above ∼210 K; consequently, their contribu-
tions to chlorine activation at midlatitudes is minimal.6
Burley and Johnston7 suggested that nitrosyl sulfuric acid
(NSA, formula NOSO4H) could be readily formed in sulfuric
acid aerosol in the stratosphere and that reaction of NSA with
dissolved HCl (or Cl-) could yield ClNO. ClNO, once in the
gas phase, is readily photolyzed; thus, the proposed scheme
results in activation of chlorine. The formation of NSA is
favored in more concentrated sulfuric acid (thus warm atmo-
spheric temperatures), and reaction of HCl with NSA might be
a source of active chlorine at midlatitudes. Burley and Johnston
pointed out that the N(III) compound of NSA may be present
as H2ONO+ (i.e., protonated HONO). The overall reaction can
be written as follows:
onto 60-72 wt % sulfuric acid solutions doped with HONO.
Although their measurements (pHCl ) pHONO ) 3-5 × 10-7
Torr) were not done under stratospheric conditions, Zhang et
al. estimated that the production of ClNO could be significant,
especially under high volcanically perturbed aerosol conditions.
In their calculations estimates for HONO solubility were used.
However, Becker et al.10 have recently reported the solubility
of HONO in sulfuric acid over wide temperature and composi-
tion ranges, which necessitates revising Zhang et al.’s estimates.
Using a Knudsen cell, Fenter et al.11 also studied the reaction
of HONO with HCl on sulfuric acid. Interestingly, they
observed no reaction between HCl and HONO on solutions
containing more than 65 wt % sulfuric acid. At 60 wt % they
observed a maximum uptake coefficient for HCl in the presence
of HONO of 2 × 10-3. Because they did not observe ClNO
formation at higher wt %, they suggest that NSA is not an
important intermediate in this reaction and that this reaction
will have a limited impact on the volcanically unperturbed
stratosphere.
Because of the discrepancies between the experiments of
Zhang et al. and Fenter et al., reinvestigation of this reaction
seemed prudent. By measuring the reaction probability (γ),
which is composed of fundamental physicochemical param-
eters,12 and estimating the heterogeneous processing of this
reaction, it should be possible to determine its stratospheric
significance. The reaction probability on a sphere of radius a
is related to the mass accommodation coefficient (R) and the
bulk liquid-phase parameters by the equation:6
HONO + HCl S ClNO + H2O
(1,-1)
The gas-phase enthalpy change for reaction 1 with HONO
in the trans form is -4.5 kcal/mol.8 With a calculated standard
entropy change of 3.4 cal mol-1 K-1,8 the gas-phase free energy
change (∆G°) at 298 K is approximately -5.5 kcal/mol.
Zhang et al.9 reported measurements of HONO and HCl
reactive uptake in a coated wall flow tube and concluded this
1
γ
1
R
ω
)
+
(2)
4RTH k1D f(a/l)
x
1
† Permanent address: National Institute for Environmental Studies, 16-2
Onogawa, Tsukuba, Ibaraki 305 Japan.
where ω is the mean thermal speed, H is the Henry’s Law
coefficient, kI is the first-order loss rate coefficient (kI )
k11HHClp(HCl)), Dl is the liquid-phase diffusion coefficient, and
l the reacto-diffusive length, l ) (Dl/kI)1/2. f(a/l) represents the
‡ Also affiliated with the Department of Chemistry and Biochemistry,
University of Colorado, Boulder, CO 80309.
* To whom correspondence should be addressed.
S1089-5639(98)00712-9 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/18/1998

3324 J. Phys. Chem. A, Vol. 102, No. 19, 1998
Longfellow et al.
function [coth(a/l)-l/a] and for bulk measurements or, if a .
l, this factor is equal to unity. From reactive uptake measure-
ments of γ, the quantity H(kIDl)1/2 can be determined and, if H
and Dl are known or can be estimated, values for kI can be
obtained. If eq 2 applies and if all these parameters are known
or can be estimated, the reaction probability for reaction 1 can
be calculated for application to the atmosphere.6 In this way,
although laboratory measurements of the reaction probability
of HONO and HCl are taken under conditions unlike those in
the stratosphere, uptake coefficients applicable to the strato-
sphere can be obtained.
Here we describe experiments on bulk liquids and on aerosol
particles that elucidate the reactions of HONO in sulfuric acid
solutions. The solubility of HONO in sulfuric acid for a number
of different H2SO4 concentrations was measured. The reaction
of HCl with HONO was investigated on 50, 60, 67, and 70 wt
% sulfuric acid solutions. We show that the deviation of the
data from bulk phase reaction behavior necessitates including
a surface specific term in the equation used to describe the
reactive uptake. The hydrolysis of ClNO (-1) on sulfuric acid
solutions was also studied to further understand the surface
specific behavior. Measurements of HONO and ClNO uptake
onto sulfuric acid particles are presented and they indicate the
need to consider a surface reaction to describe γ. The reaction
of HONO with saturated NaCl solutions was investigated to
determine if the presence of protons is necessary for this reaction
to take place.
Briefly, a known amount of gaseous HONO was absorbed into
a known volume of solution (stirred continuously), thus resulting
in an absolute measure of the Henry’s law coefficient (HHONO).
A second type of absolute measurement was performed wherein
a nitrosyl-sulfuric acid solution was prepared by adding a
known amount of NaNO2 to 80 wt % H2SO4 at 0 °C. The
composition of this solution was ∼0.4 M NaNO2 in ∼79 wt %
H2SO4. The vapor pressure of HONO over these solutions was
monitored over a wide temperature range to derive HHONO
.
Reactive Uptake Measurements. HONO reactive uptake
was studied on 50 (205 and 250 K) and 60 (219, 230, and 250
K) wt % solutions. For the HONO reactive uptake experiments,
HCl was introduced into the reactor with the He carrier gas
(pHCl, the partial pressure of HCl, was between 10-10 and 10-7
atm) and allowed to equilibrate with the liquid. Concentration
of HCl in the liquid was between ∼5 × 10-6 and ∼3 × 10-2
M, dependent on both [H2SO4] and temperature (the Henry’s
law coefficients for HCl were taken from Carslaw et al.16). The
RWW was primarily used in these experiments. HCl reactive
uptake onto 67 (250K) and 71 (215 and 269 K) wt % sulfuric
acid solutions that were doped with HONO was studied. HONO
was added to the liquid either through the gas phase by including
HONO with the carrier gas, feasible when HHONO is relatively
low (<105 M atm-1), or HONO/H2SO4/H2O solutions were
prepared by mixing known amounts of NaNO2 with sulfuric
acid solutions. ClNO uptake onto 60 wt % sulfuric acid,
reaction (-1), was also studied at 250 K. A ∼1% mixture of
ClNO (99% from Matheson) in He was used. Possible
impurities such as Cl2 or NO in the ClNO were not observed.
Even if present these species are not readily taken up by sulfuric
acid and would not participate in the observed heterogeneous
reaction. For the uptake of HONO on NaCl, a saturated NaCl
solution was prepared by placing dry NaCl crystals in the flow
tube at 268 K and then adding water vapor to the He flow until
the salt appeared wet (deliquesced). The reactive uptake of
HONO was then studied.
Experimental Section
The uptake in and solubility on bulk liquids was measured
in a cylindrical flow tube reactor. The liquid sulfuric acid
solutions (50-83 wt % H2SO4) were located on the inside wall
of the flow tube (i.d. ) 2.2 cm)13 or inside a rotating wetted
wall (RWW, i.d. ) 1.84 cm) flow reactor, recently developed
in our laboratory.14 For the RWW experiments, the liquid is
located on the inside of a glass cylinder that is rotated (0-10
rpm) to maintain a cylindrical coated wall liquid surface. A
few experiments were performed with the liquid located in a
small container at the bottom of the flow tube.13
The average velocity of the He carrier gas (typically 1000
cm s-1) was between 200 and 2000 cm s-1, the total pressure
(typically 1 Torr) was between 0.5 and 5 Torr, and the
temperature ranged from 205 to 295 K. The standard analysis
procedures for flow tubes were used to calculate the reactive
uptake coefficients.17,18 Corrections to the observed first-order
rate coefficients due to diffusion ranged from a few percent to
as much as 50%. The diffusion coefficients for HONO in He
and in H2O were calculated to be 290 and 75 Torr cm2 s-1 at
220 K, respectively.19-21
Solubility Measurements. Because the Henry’s Law coef-
ficients for HONO (HHONO) varied over a large range, a few
different techniques were employed to measure them. Solubili-
ties were measured using both relative and absolute methods.
Solubilities reported here include physical uptake as well as
uptake resulting from any further reaction in solution, such as
protonation or dissociation, and the values obtained are com-
monly referred to as effective Henry’s Law coefficients.
Relative HONO solubilities were measured by monitoring
the partial pressure of HONO over a HONO-doped sulfuric acid
solution and varying either the temperature or the water content
of the solution. The solutions were prepared by adding HONO
from the gas phase to a small amount of solution located in the
RWW. HONO was synthesized by adding small amounts of
dilute NaNO2 solution to 20 wt % sulfuric acid at 0 °C and
transporting the resulting HONO vapor to the flow tube through
a movable injector. The solutions in this case were continuously
stirred by the slowly rotating RWW and aided by a glass
encapsulated stirring bar (∼10 cm × 0.5 cm o.d.) that rested in
the liquid. The HONO signal (proportional to partial pressure)
was then monitored as a function of temperature or of water
partial pressure. Care was taken to minimize the amount of
HONO desorbed from the solutions during the measurements
(i.e., concentration of HONO in the substrate remained constant).
Absolute HONO solubilities were determined in 60 and 65
wt % H2SO4 solutions using the RWW flow tube method.15
Gas-phase species (HONO, HCl, and ClNO) were detected
using chemical ionization mass spectrometry (CIMS). Within
the CIMS, SF6- reacts with HCl22 and HONO (reaction 3a) via
-
flouride transfer. SF6 can also transfer a fluorine atom
(reaction 3b); the products in this case are ambiguous, but could
be HF + NO2.
HONO + SF6- f HF‚NO2- + SF5
f SF5- + other products
(3a)
(3b)
The rate coefficients for reaction 3 have not been reported.
-
Similar to the reaction of SF6 with HCl (and unlike the case
-
for SF6 + HNO3), it was found that fluoride ion transfer is
-
not the main channel for reaction of SF6 with HONO:
-
primarily SF5 is produced. Channel 3b accounts for ∼75%
of the products, measured by comparing the product ion signals
and assuming negligible mass discrimination in the quadrupole

HONO Solubility and Reactivity
J. Phys. Chem. A, Vol. 102, No. 19, 1998 3325
mass spectrometer. This differs from the work of Zhang et al.9
where it was assumed that flouride ion transfer is dominant.
They assumed that the rate coefficients for the reactions of
-
-
-
HNO3 and HONO with SF6 to give HFNO3 and HFNO2
,
respectively, were the same to obtain an estimate for their
HONO concentrations; our results show that this is incorrect,
as discussed below. Reaction 4 was studied previously by
Streit23 and was reported to have a rate coefficient of 4.4 ×
10-11 cm3 molecule-1 s-1 at 298 K. Our calibrations suggest
that this value is too low, see below.
ClNO + SF6- f Cl- + SF6 + NO
(4)
By monitoring the signal at HFNO2-, Cl-, and HFCl-, the
reactants and the ClNO product from reaction 1 can be
unambiguously identified.
The rate coefficients for reactions 3 and 4 can be estimated
(relative to that for HCl + SF6-) from the product ion signals
where the reaction of ClNO with H2O, reaction (-1), was
observed, and HCl and HONO were assumed to be produced
in equal amounts. Because the CIMS signals may depend on
the mass of the product ion (i.e., mass discrimination in the
instrument) as well as the ion-molecule rate coefficient,22 the
relative rate coefficients are reported as estimates. Note that
the CIMS sensitivities for HONO and ClNO, however, are as
accurate as for the HCl calibrations, ∼25%, subject to the
assumption of unity yield for reaction (-1). From the ratio of
the product ion signals, a rate coefficient of ∼6 × 10-10 cm3
molecule-1 s-1 is estimated for reaction 3a resulting in an overall
rate coefficient for reaction (3a+3b) of ∼2 × 10-9 cm3
molecule-1 s-1. By comparing the decrease in the Cl- signal
to the increase in the HFCl- signal, the estimated rate coefficient
for (4) is ∼2 × 10-10 cm3 molecule-1 s-1, significantly different
than the previously reported value of 4.4 × 10-11 cm3
Figure 1. Effective HHONO as a function of water pressure at 269 (open
circles) and 295 K (filled circles). The symbols with a cross indicate
data taken when the water vapor was increased and agree quite well
with data taken when the water vapor was decreased (plain symbols).
The lines shown have no physical meaning and are only a visual aid.
molecule reaction time calculated from the CIMS flow tube
interaction length (50 cm) and the flow rate (corrected for
concentration gradients in a laminar flow17).
Reactions on Aerosol Particles. Reactions 1 and (-1) were
also studied on sulfuric acid aerosol particles, primarily at 250
K. HONO and HCl were prepared, delivered, and detected as
described above. In separate experiments, ClNO was added to
the flow tube via the movable injector from a ∼1% mixture in
N2. Sulfuric acid particles were generated by flowing N2 over
hot sulfuric acid after which the particles flowed through a
conditioner containing 60 wt % sulfuric acid. The particle
number density was measured in an expansion type particle
counter. The size distribution was determined from the ap-
plication of Mie theory to the measured attenuation of UV light
scattering by the particles. Detailed experimental descriptions
and procedures have been given previously.24,25
molecule-1 s-1 23
.
Note also that the rate coefficient we used
for the reference reaction of HCl + SF6 (1.5 × 10-9 cm3
molecule-1 s-1 from ref 22) differs substantially from that
measured by Streit (4.2 × 10-10 cm3 molecule-1 s-1).
Using the measured CIMS sensitivity, [HONO] can be
estimated:
-
[HONO] ) (S66/S146)/kt
(5)
-
where (S66/S146) is the ratio of the signals at HF‚NO2 (m/e )
Results
-
66) and SF6 (m/e ) 146), k is the rate coefficient for that
channel (channel 3a), and t is the average ion molecule reaction
time (typically ∼5 ms). This estimate is usually accurate to a
factor of 2 (+100/-50%), but can be +30/-25% when t is
known from calibrations (see below). The initial concentration
of HONO in the reactive uptake experiments was in the range
of 1010 to 1011 molecule cm-3. For measuring the gas-phase
HONO over the prepared samples of NSA in sulfuric acid and
for the HCl reactive uptake measurements, eq 5 was used. For
the absolute solubility measurements using the RWW method15
or the HONO reactive uptake measurements, it is only necessary
to know the relative changes in the HONO gas-phase concentra-
tions.
The CIMS was calibrated for HCl by adding a known flow
(∆P/∆t was measured in a known volume) of a dilute HCl
mixture to the flow tube. Both a 0.3% HCl-in-He mixture and
a 1.0% HCl-in-N2 mixture were used; the HCl content of the
mixtures were known by manometric preparation and were
checked by absorption at 185 nm. With the known rate
coefficient for SF6- + HCl and an analogous equation to eq 5,
t can be calculated. It was usually within 30% of the ion
Relative Henry’s Law Coefficients. Shown in Figure 1 are
the results of an experiment where the water content of the
solutions was changed by varying the partial pressure of water
at a constant temperature. Relative HHONO values, which are
inversely proportional to the HONO signal and thus partial
pressure, were normalized to HHONO determined from absolute
measurements (see below). HHONO is strongly dependent on
water partial pressure (and thus H2SO4 content) of the solutions.
In these experiments, the H2O vapor pressure, calculated from
mass flow measurements (see note 23 in Hanson and Lovejoy15),
was initially 0.1 and 0.7 Torr at 269 and 295 K, respectively.
Then the solutions were dried by flowing He over them, a lower
H2O partial pressure was set, and the HONO signal recorded.
This process was reversed (i.e., H2O was added), and the results
obtained from drying out and adding H2O were in excellent
agreement. Variations in the volume of the solutions due to
changes in density and water content were taken into account
when calculating relative Henry’s law coefficients. The density
of the sulfuric acid solutions was taken from Huthwelker et al.26
The Henry’s law constants determined from these measurements

3326 J. Phys. Chem. A, Vol. 102, No. 19, 1998
Longfellow et al.
TABLE 1: Measured Solubility Values
acid, Henry's law coeff,
T, K wt %
M atm-1
method
relative
uncertainty
295
295
295
295
292
280
274
273
261
250
230
269
295
269
269
295
271
269
249
230
269
295
270
296
270
270
249
240
220
240
220
218
83
82
79
79
79
79
79
79
79
79
79
77
76.5
74
72
71
71
71
71
71
70.5
70
70
66
66
65
65
65
65
60
60
60
1.3 × 107
6.6 × 106
9.1 × 105
1.5 × 106
3.1 × 106
8.5 × 106
1.6 × 107
1.6 × 107
7.7 × 107
2.8 × 108
4.6 × 109
1.7 × 107
2.6 × 105
1.3 × 106
1.4 × 105
9.8 × 103
9.3 × 104
9.1 × 104
1.1 × 106
1.2 × 107
5.9 × 104
6.3 × 103
5.8 × 104
1.2 × 103
4.2 × 103
2.6 × 103
9.0 × 103
2.4 × 104
1.6 × 105
3.0 × 103
2.7 × 104
4.1 × 104
+100/- 50%
+100/- 50%
+100/- 50%
relative
relative
abs. molaritya +150/-65%
abs. molarity
abs. molarity
abs. molarity
abs. molarity
abs. molarity
abs. molarity
abs. molarity
relative
relative
relative
relative
relative
+150/-65%
+150/-65%
+150/-65%
+150/-65%
+150/-65%
+150/-65%
+150/-65%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+100/- 50%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
+30/-25%
relative
relative
relative
relative
Figure 3. A second type of absolute Henry’s law measurement in
which the HONO concentration in solution is known and the HONO
pressure is monitored as a function of temperature. The HONO
concentration is ∼0.4 M in 79 wt % sulfuric acid.
relative
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
abs. RWW
a abs. molarity ) absolute molarity.
Figure 4. Plot of the raw data (open symbols) from the loss of HONO
on 50 and 60 wt % sulfuric acid solutions with HCl at 250 K. The
solid lines are a linear fit through the data up to an injector position of
6 (11 for 60 wt %) cm. The effects of equilibrium become quite
pronounced at injector positions greater than 6 for the 50 wt % data.
The data points are corrected for the production of HONO from ClNO
(solid symbols) for both 50 and 60 wt %. The dashed lines are the fit
through the corrected data and γ was calculated from these corrected
slopes. See text for details.
K down to 230 K and then back to 296 K with good
reproducibility in the vapor pressure of HONO. HHONO in 79
wt % H2SO4 can be obtained from these data and these values
are given in Table 1 (absolute molarity measurements). A small
correction to the [HONO] in solution from the increase in
solution density as the temperature drops (∼5% from 295 to
230 K) was applied.
Reactive Uptake of HONO on HCl-Doped Solutions. The
reaction probabilities for HONO were measured over a range
of HCl partial pressures. No reaction between HONO and HCl
was observed without sulfuric acid present (i.e., no reaction in
the gas phase or on glass). Figure 4 shows the data for HONO
uptake on 50 and 60 wt % sulfuric acid at 250 K. As the
reaction is monitored over longer contact times, the observed
loss of HONO (at 50 wt % especially) slows for injector
positions larger than ∼6 cm. The reaction begins to reach
equilibrium (i.e., [ClNO][H2O]/[HCl][HONO] approaches Keq
Figure 2. HONO gas-phase signal measured by the CIMS as a function
of time. By monitoring the absorption or desorption of HONO into a
known amount of sulfuric acid an absolute solubility can be measured.
after normalization are given in Table 1 (relative measurements).
Absolute Henry’s Law Coefficients. The physical uptake
of HONO onto 60 and 65 wt % H2SO4 solutions was
investigated over a range of temperatures (218-273 K). Data
for the uptake/solvation of HONO into ∼0.44 cm3 of 65 wt %
H2SO4 at 249.5 K are shown in Figure 2. Both the uptake and
desorption of HONO are shown and the amount of HONO
absorbed is equal to that which is desorbed (within 10%).
Results for a number of such experiments are presented in Table
1 (absolute RWW measurements).
Shown in Figure 3 are the measured partial pressures of
HONO over the 0.4 M HONO in 79 wt % H2SO4 solution as a
function of 1/T. The solution temperature was varied from 295

HONO Solubility and Reactivity
J. Phys. Chem. A, Vol. 102, No. 19, 1998 3327
) exp(-∆G°/RT)) and the hydrolysis of ClNO results in
significant production of HONO. The effect at 50 wt % sulfuric
acid is thought to be more pronounced because of higher [H2O],
which increases the rate of the reaction in the reverse direction.
To account for this equilibrium effect as fully as possible,
the raw data points were corrected for the amount of HONO
produced from the reaction of ClNO with H2O. This was done
by taking into account the amount of HONO produced from
ClNO at each injector position, the uptake coefficient for ClNO
(based on experiments discussed below), and the average
reaction time (injector position/average flow velocity.) In
addition, the fraction of this HONO produced that reacted with
HCl was calculated.27 As illustrated in Figure 4, this correction
can be significant. The data shown is typical for the 50 wt %
measurements, and without correcting for HONO production
from ClNO γ would by underestimated by ∼40%. Most
measurements did not require such a large correction; for
example, the uncorrected 60 wt % data would yield a 20% lower
γ on average. Neither Zhang et al.9 or Fenter et al.11 reported
the occurrence of the reverse reaction in their experiments. We
found that the effects of reverse reaction (-1) could be obscured
if the reaction was not taking place under pseudo-first-order
conditions in HONO (i.e., [HCl] in solution was not at least 6
times greater than the [HONO]).
The dependence of γ on HCl pressure for 50 wt % sulfuric
acid solutions at 205 and 250 K is illustrated by a log(γ)-log-
(pHCl) plot (Figure 5a). The slope of the linear fit to the 50
wt % data at 205 K is 0.53 ( 0.08. The uncertainty here, as
well as in all the log(γ)-log(pHCl) measurements, represents
2σ in the fit. γHONO was weighted in the fits by uncertainties
of one-half the difference between the uncorrected value and
the value corrected for HONO production from ClNO (averaged
( 12% for the 50 wt % data at 205 K). The value of 0.53 is
close to the value of 0.5 expected from eq 2 indicating that
HONO uptake can be described using a bulk reaction with HCl
at 205 K. However, at 250 K the slope is 0.80 ( 0.22,
indicating that the reaction of HONO with HCl cannot be
completely described as a bulk reaction (γ uncertainties averaged
(38%). Alternatively, the same data can be plotted as Γ versus
(pHCl)1/2 where 1/Γ ) 1/γ - 1/R. The data plotted in this
manner should lie on a straight line if the reaction probability
can be described by bulk processes and the mass accommodation
coefficient is ∼1.
Figure 5. Data taken at 50 wt % and 205 and 250 K. (a) Log(γ)-
log(pHCl) plot illustrating the relationship between HCl pressure and
the uptake coefficient. (b) The fit to the data at 205 K using eq 2 (solid
line) is nearly identical to that from eq 8 (dashed line). (c) It becomes
clear in this plot that eq 2 is not adequate to completely explain the
reaction taking place at 250 K. By including a surface contribution
(dashed line) the data is reproduced significantly better.
Equation 7 (and thus 8) is an approximation29 which, for the
results presented here (i.e., γ e 0.1), can be considered a good
one. The dashed line in Figure 5b is a second-order fit to γ
versus (pHCl)1/2 according to eq 7. Values for a1 and a2 (eq 8)
are given in Table 2 for all the measurements where a1 is related
to the liquid-phase second-order rate coefficient (kII) and a2 is
4RTHHONO kIIHHCl(pHCl)Dl
-1
x
related to a surface specific reaction rate coefficient (k
M ).
sur
Γ )
(6)
Values for kII were obtained by calculating HHCl and using
the HHONO measured in this work or, for 50 wt %, that of Becker
et al.10 Dl for HONO was calculated by assuming the same
parameters that were used for HOCl.26
16
ω
As expected, a linear fit of Γ vs (pHCl)1/2 represents the 205
K data reasonably well (Figure 5b), but does not describe the
250 K data as depicted in Figure 5c. Since γ for HONO is not
linearly dependent on (pHCl)1/2 in 250 K sulfuric acid, the
reaction of HONO with HCl likely has a contribution resulting
from a surface specific reaction on sulfuric acid. In these
experiments, R . γ and the curved behavior cannot be attributed
to γ f R. In addition, varying R from 0.1 to 1 does not effect
this behavior.
At 60 wt % it was necessary to use only measurements
wherein the HCl concentration in the solution phase was at least
a factor of 6 higher than the HONO concentration. This is
because HHONO is larger than HHCl at 60 wt % and higher [HCl]
than [HONO] in the liquid phase is necessary to ensure first-
order conditions in HONO. Figure 6 consists of (a) a log(γ)-
log(pHCl) plot and (b) a Γ vs pHCl1/2 plot, comparing the data
at 219, 230, and 250 K. The results and parameters used for
both 50 and 60 wt % are also compiled in Table 2. The 2σ
uncertainties in the fits to a1, and therefore in kII, are given in
Table 2 and average (56%. This does not represent the total
A surface reaction term, which is proportional to the HCl
content of the liquid (HHCl(pHCl)), can be added to eq 6.28,29
4RTHHONO kIIHHCl(pHCl)D
x
uncertainty in calculating kII as uncertainties in HHCl, HHONO
,
Γ )
l + ksurHHCl(pHCl) (7)
(8)
ω
and Dl are not included. Most of the measurements exhibit
surface specific behavior except at 60 wt %, 219 K and 50 wt
%, 205 K where the reaction occurs primarily in the bulk.
x
Γ ) a1 pHCl + a2(pHCl)

3328 J. Phys. Chem. A, Vol. 102, No. 19, 1998
Longfellow et al.
TABLE 2: Uptake Measurements
acid,
wt %
log-log
slope ( 2σ
HHONO
,
HHCl
,
Dl,
kII,
T, K
M atm-1
M atm-1
cm2 s-1
M-1 s-1
a1 ( 2σ
a2
HONO Uptake
250
205
250
230
219
50
50
60
60
60
0.80 ( 0.22
0.53 ( 0.08
0.78 ( 0.10
0.66 ( 0.06
0.47 ( 0.16
3.0 × 102
4.2 × 104
1.3 × 103
1.0 × 104
4.0 × 104
2.5 × 104
3.0 × 106
8.6 × 102
4.7 × 103
1.4 × 104
9.5 × 10-7
3.6 × 10-8
6.3 × 10-7
1.9 × 10-7
7.1 × 10-8
8.1 × 103
1.5 × 103
9.4 × 104
6.9 × 104
5.0 × 104
10 ( 11
1200 ( 250
23 ( 11
8.9 × 104
3.3 × 106
1.9 × 105
8.3 × 105
0
180 ( 38
640 ( 530
HCl Uptake
269
215
250
70
70
67
0.92 ( 0.24
0.46 ( 0.10
0.90 ( 0.18
1.5 × 105
1.0 × 108
∼7 × 104
5.7
2.3 × 102
60
9.6 × 10-7
1.7 × 10-8
8.0 × 10-7
5.8 × 104
3.5 × 105
3.9 × 104
1.2 ( 3.9
350 ( 25
6.9 ( 11
8.7 × 104
0
3.8 × 105
Figure 7. A plot of measured reactive uptake coefficients versus
particle size at 250 K, 60 wt %, and two HCl pressures. The solid line
is a fit of the data to eq 9 at 1.2 × 10-7 atm allowing both l and Γ to
vary. For the high-pressure data the short dashed line indicates the fit
while the upper limit to l is also plotted (dotted line).
here as variable to be fit, as is l. Here we assume no surface
contribution to illustrate that the measured γ on particles can
be directly related to the reacto-diffusive length.
From the second-order rate coefficient (k11) that was obtained
from the 60 wt % bulk data at 250 K, a reacto-diffusive length
of 2.6 µm is predicted for an HCl pressure of 1 × 10-7 atm.
However, as evident in Figure 7, a reacto-diffusive length of
0.04 ( 0.04 ((2σ) µm can represent the data. A similar trend
is also observed at higher HCl pressures where an l value of
1.3 µm is predicted, yet the data indicates an l value of
essentially zero (l < 0.02). Since the reacto-diffusive length
can be considered a measure of the depth over which the reaction
is taking place, this small value is an indication that a surface
reaction is dominating the conversion. Note that these measure-
ments do not yield a value for l for the bulk phase reaction;
rather they should be considered further evidence that this
reaction has a surface component.
Reactive Uptake of HCl on HONO-Doped Solutions. HCl
loss measurements onto ∼70 wt % sulfuric acid solutions doped
with HONO were measured at 215 and 269 K. Reactive uptake
measurements on 67 wt % solutions at 250 K were also carried
out. It was necessary to dope the solutions with HONO and
follow the HCl uptake as the large HHONO for these sulfuric
acid compositions (HHONO . HHCl) precludes pseudo-first-order
conditions for HONO in the liquid. These data was not
corrected for HCl production from reaction (-1) as the quantity
[ClNO][H2O]/[HONO][HCl] was far from the equilibrium value
(Keq ) exp(-∆G°/RT)). Uncertainties in γ averaged (15%.
Figure 6. Data from 60 wt % solutions at 219, 230, and 250 K are
compared. Figure 6a is similar to Figure 5a. The solid lines in Figure
6b are fits using eq 8 and assuming a surface specific reaction.
Reaction of HONO on HCl-Doped Sulfuric Acid Aerosol.
The reactive uptake of HONO onto HCl-doped 60 wt % sulfuric
acid aerosol particles was studied as a function of particle size
at 250 K. The production of HONO from the reverse reaction
(-1) was taken into account, as in the bulk reactions. In Figure
7 the measured γ from the aerosol experiments is plotted against
rs, the surface-area weighted particle radius.14 rs is a good
approximation (for log-normal distributions with geometric
standard deviations e1.26)14 that takes into account the poly-
disperse aerosol samples that we use. No significant size
dependence for γ was observed in the pHCl range from 1 to 5
× 10-7 atm. The lines are fits to the data according to eq 2,
which rearranges to eq 9. Γ, defined by either eq 6 or eq 7
RΓf(rs/l)
γ )
= [coth(rs/l) - l/rs]Γ
(9)
Figure 8a is a log(γ)-log(pHONO) plot for three different
conditions. At 250 and 269 K the slope is close to 1 (0.90 (
0.18 and 0.92 ( 0.24, respectively), while at 215 K and 70 wt
R + Γf(rs/l)
depending on whether a surface contribution is present, is treated

HONO Solubility and Reactivity
J. Phys. Chem. A, Vol. 102, No. 19, 1998 3329
Figure 9. ClNO uptake onto 60 wt % sulfuric acid aerosols at 250
and 294 K as a function of the surface area weighted particle radius.
This reaction also shows the effect of a surface reaction with small l
values necessary to fit the data.
HONO Uptake onto NaCl Solutions. No reaction of HONO
with Cl- to give ClNO was observed over saturated NaCl
solutions (∼6 M) at 268 K. The signal due to ClNO showed
no increase; however, in some experiments, production of
HONO was noticed. HONO is not produced exclusively in the
source. NO2, NO, and H2O are in equilibrium with HONO as
shown in reaction 10.31
2HONO S H2O + NO + NO2
(10,-10)
Figure 8. (a) Log(γ)-log(pHONO) plots of 70 and 67 wt % data at
215, 269, and 250 K. (b) A plot of Γ versus p(HONO)1/2 allows kII to
be extracted.
Either reaction (-10) or reaction 11 could lead to HONO
production if [H2O] is significantly different in the flow tube
than in the HONO source.
% the data can be fit with a slope of 0.46 ( 0.10. Figure 8b is
a plot of Γ(HCl) versus (pHONO)1/2 for these different condi-
tions. Similar to the HONO uptake experiments, a second-order
rate coefficient (kII) can be extracted and these values are listed
in Table 2. Dl for HCl was calculated analogously to that for
HONO with slight modifications for HCl.30 The large uncer-
tainty in a1 (Table 2) at 269 K reflects the fact that these data
could be fit without a contribution from the bulk reaction. This
uncertainty is also reflected in the kII values. Not included are
uncertainties due to the Henry’s law coefficients and Dl as
discussed earlier.
2NO2 + H2O f HONO + HNO3
(11)
The high [H2O] in the flow tube, which was necessary to saturate
the NaCl, probably led to the HONO production. In the
experiments over sulfuric acid, the water vapor pressure was
much lower and production of HONO was not observed. These
experiments over saturated NaCl solutions indicate that the
reaction of HONO with Cl- requires the presence of protons to
be efficient.
Reverse Reaction: ClNO Hydrolysis. The reaction prob-
ability for loss of ClNO due to hydrolysis (-1) was investigated
at 250 K on 50 and 60 wt % sulfuric acid in the RWW. A
small correction for the effect of reaction 1 on the observed
ClNO loss rate coefficient was made. There was no evidence
of impurities in the ClNO affecting the uptake. The average
reaction probability for ClNO was (2.5 ( 0.8) × 10-3 on 60
wt % and (3 ( 1) × 10-3 on 50 wt % H2SO4 at 250 K. Without
making assumptions about ClNO solubility, which would be
unreliable, or postulating that no surface-specific reaction is
occurring, which would be incorrect, it is not possible to
determine the bulk hydrolysis rate coefficients for ClNO.
The reaction probability for loss of ClNO due to hydrolysis
was also investigated as a function of particle size at 250 and
294 K on 60 wt % sulfuric acid aerosol particles. These results
are plotted in Figure 9 and no significant dependence on size
was found (l e 0.02 µm at both temperatures). The reaction
Discussion
HONO Solubility. The solubility of HONO in H2SO4
solutions is a strong function of acid content. As found by
Becker et al.,10 HONO is much more soluble in strong acid
(∼60-70 wt % H2SO4) than in more dilute H2SO4 (∼50 wt
%). Plotted in Figure 10 is the Henry’s law coefficient (M
atm-1) versus H2SO4 content for 295 K (open symbols) and
269 K (filled symbols) (both relative and absolute). The H2-
SO4 content was calculated from the water partial pressure and
temperature for the data presented in Figure 1.32 To apply all
the data in Table 1 to this figure, some of the data was
“corrected” for a small temperature difference (∆T < 3 K) using
a ∆H° of -15 kcal mol-1. Also shown as the triangles are
sol
the data of Becker et al.10 corrected for [HONO] reported in
mol kg-1 solution-1 to mol L-1 (small T corrections were also
made to some of these data). There is excellent agreement
between the results of the present work and the results of Becker
et al. Our results extend the range over which HHONO has been
measured to 79 wt % H2SO4 and to temperatures as low as 220
K.
probability at 250 K is in good agreement (γ ) 2.5 × 10-3
)
with that measured over bulk solutions here. The average γ
measured on particles at 294 K, 5.0 × 10-3, indicates that γClNO
is temperature dependent.

3330 J. Phys. Chem. A, Vol. 102, No. 19, 1998
Longfellow et al.
solvation enthalpies and include solvation as well as protonation,
dissociation, or association. The difference in ∆Hsol at 60 and
65 wt % as compared to 71 and 79 wt % most likely indicates
a change in the form of HONO in solution. Becker et al.
observed a dramatic change in the variation of HHONO with H2-
SO4 content near 53 wt %, which they attributed to the formation
of NO+ via the equilibrium reaction 12.
HONO + H+ S H2O + NO+
(12)
However, there are other possibilities, such as H2ONO+,
which has also been identified by spectroscopic techniques in
solution.7 From the strong evidence of a species shift at 53 wt
% in Becker et al.’s work, we agree that the form of N(III) in
solution is HONO below 53 wt %. However, from our effective
solvation values we suspect that the equilibrium is more
complicated than as represented by eq 12. It is possible that
primarily H2ONO+ is the N(III) species in 60 and 65 wt %,
-
and a mixture of H2ONO+ and NO+HSO4 is present in 71
and 79 wt % solutions.34
Atmospheric Implications. The amount of N(III) (i.e.,
HONO, H2ONO+, and NO+HSO4-) found in stratospheric
aerosols can be estimated from the results of this work assuming
HONO is in gas-liquid equilibrium. Here we use a gas-phase
HONO concentration of 1 × 107 molecule cm-3, a relatively
high estimate for the unpolluted stratosphere;35 this is equivalent
to a HONO partial pressure of ∼3 × 10-13 atm. The
concentration of N(III) in 79 wt % sulfuric acid at 232 K would
be ∼1.2 × 10-3 M using our measured Henry’s law constant.
At 71 wt % and 215 K a sulfuric acid aerosol would contain
∼3 × 10-5 M N(III), and in a 60 wt % aerosol at 202 K the
concentration would be ∼9 × 10-8 M. Even at high HONO
concentrations, as measured in airplane plumes (0.52 ppbv ∼2
× 109 molecule cm-3),35 the concentration of N(III) in solution
would be 0.23 M, 5.9 × 10-3, and 1.8 × 10-5 at 79, 71, and 60
wt %, respectively. Burley and Johnston calculated that N(III)
could be as high as 1% of the sulfuric acid solution or ∼0.15
M in a 60 wt % aerosol; this could not result from HONO uptake
unless HONO was on the order of 5 ppmv.
Figure 10. Effective HHONO values as a function of sulfuric acid wt %
at 269 (open symbols) and 295 K (filled symbols). The circles represent
relative measurements from this work, while the squares are absolute
measurements from this work. The triangles are measurements from
Becker et al.
To extrapolate our measured γ values to the stratosphere it
is necessary to know kII, HHONO, HHCl, and Dl. Assuming an
HCl concentration of 2 × 109 molecule/cm3, a γstrat for HONO
uptake can be calculated using the following eq 13, which differs
4HHONORT klD
x
Γcalc
≡
lf(rs/l) + ksurHHCl(pHCl) (13)
Figure 11. Plot of effective Henry’s law constants as function of 1/T.
The open squares are absolute measurements at 79 wt %, analogous to
Figure 3. The open circles are relative measurements while the RWW
absolute measurements are represented by open triangles. The filled
squares are from Becker et al.’s measurements.10 Extrapolating to
stratospheric conditions (pH2O ) 2 × 10-4 Torr), HHONO is 3 × 105 M
atm-1 in 60 wt % acid at 202 K increasing to 9 × 105, 1 × 108, and
4 × 109 M atm-1 in 65, 71, and 79 wt % (205, 215, 232 K, respectively).
The dashed lines on either side of the diamond symbols, which represent
stratospheric conditions, illustrate the variation of HHONO as the water
vapor pressure varies from 1 × 10-4 to 4 × 10-4 Torr.
ω
from eq 7 only by including the f(rs/l) term.
l is equal to(Dl/kI)1/2 and rs is taken to be 0.1 µm to represent
background sulfuric acid aerosol conditions. To estimate Γcalc
,
the reaction rate parameters were taken to be the values
measured here at the lowest temperatures. The diffusion
coefficients for HONO and for HCl were calculated as in
Huthwelker et al.26 and Luo et al.30 The HHONO were taken
from extrapolations of the measured HHONO (present work and
Becker et al.10) and HHCl were taken from the calculations of
Carslaw et al.16 These parameters and the resulting Γcalc are
shown in Table 3. Note that Γcalc differs from γ by less that
0.2%.
Figure 11 is a master plot of the measured HONO solubilities
reported here vs 1/T for several H2SO4 solutions. Also included
are some of the results of Becker et al.10 A plot of ln(H) vs
1/T results in values of ∆H°sol for HONO.33 ∆H°sol for HONO
in 79 wt % H2SO4 is -16 kcal mol-1, that for ∼71 wt % is
-15 kcal mol-1, at 65 wt % it is -10 kcal mol-1, and in 60 wt
% H2SO4 ∆H°sol is -11.5 kcal mol-1. These are effective
The processing of HCl from reaction with HONO can be
estimated from the loss of HONO in the stratosphere via
eq 1436

HONO Solubility and Reactivity
J. Phys. Chem. A, Vol. 102, No. 19, 1998 3331
TABLE 3: Calculated Reaction Probabilities for HONO + HCl at Stratospheric Conditions
acid,
wt %
kII,
ksur
,
Dl,
HHONO
,
HHCl,
T, K
M-1 s-1
M-1
cm2 s-1
M atm-1
M atm-1
Γa
195
205
215
50
60
70
1.5 × 103
5.0 × 104
3.5 × 105
1.1
0
0
9.2 ×10-9
1.6 × 10-8
1.7 × 10-8
1.8 × 105
1.0 × 108
1.0 × 107
4.8 × 104
2.3 × 102
1.6 × 10-3
3.0 × 10-4
1.5 × 10-5 b
3 × 105
a Reaction probability for HONO assuming [HCl] ) 2 × 109 molecule cm-3 and rs ) 0.1 µm. b Reaction probability for HCl assuming [HONO]
) 1 × 107 molecule cm-3 and rs ) 0.1 µm.
based on the inability of equations derived using bulk phase
assumptions to explain the data (i.e., point (i) above). In our
experiments the aerosol data adds a second piece of evidence
that there is a surface component to this reaction. Furthermore,
we do see a variation in the importance of the surface reaction
with temperature and sulfuric acid wt %. ksur (a2/HHCl for 50
and 60 wt %, a2/HHONO for 70 wt %) is expressed most at high
temperatures and in 60 wt % H2SO4, we do not know why the
surface reaction would be favored under these conditions.
There is additional evidence for the surface reaction in that
the γClNO on aerosols is independent of particle size down to
0.05 µm (l < 0.02 µm). This indicates that the reverse reaction
(-1) is also facilitated at the surface. It is possible that the
fractional HCl content on the surface could be high and cause
what appears to be a surface-specific effect for reaction 1 (l
could be small because of high HCl). In this case, the reverse
reaction of ClNO + H2O would not show the surface effect.
Because ClNO hydrolysis is probably enhanced at the surface,
it is likely that the surface reaction term for reaction 1 is due to
the special environment of the surface rather than only an
enhancement of one or both of the reactant concentrations at
the surface.
γσAω[HONO]strat
dHCl
dt
dHONO
-
) -
)
(14)
dt
4
where σA is the aerosol surface area per unit volume, σA ) 2 ×
10-8cm-1 is typical for nonvolcanic conditions, and [HONO]strat
is 1 × 107mol cm-3. The processing of HCl at 60 wt % and
205 K is 0.4 molecule cm-3 s-1, while at 50 wt % and 195 K
it is 1.5 molecule cm-3 s-1. At 70 wt % and 215 K (replace
[HONO]strat with [HCl]strat ) 2 × 109 molecule/cm3 and replace
Γ
HONO with ΓHCl ) 1.5 × 10-5), the HCl processing rate is 5.3
molecule cm-3 s-1. The major loss process for HCl in the
stratosphere at these warmer temperatures (i.e., >210 K) is
reaction with OH (typically 1 × 106 molecule cm-3 s-1), which
results in a processing rate for HCl of ∼1200 molecule cm3
s-1 for these conditions.37 These calculations indicate that
reaction with HONO (even at a relatively high concentration
of 1 × 107 mol cm-3) is not a significant loss process for HCl.
Our results differ from Fenter et al.11 in that we do observe
reaction of HONO and HCl in sulfuric acid solutions greater
than 65 wt % acid. The high values for γ observed by Zhang
et al.9 (0.01-0.02 for 60-72 wt %) were not measured under
stratospheric conditions; their reactant concentrations of 3-5
× 10-7 Torr (∼1 × 1010 molecule/cm3 at 215 K) are higher
than what would be found in the stratosphere, especially for
HONO. If γ is calculated for their conditions, similar to the
process used in Table 3, a value of ∼5 × 10-3 is obtained at
220 K for 70 wt % solutions; this differs somewhat from their
measurement of 1.9 × 10-2. Their ClNO production rate of
34 molecule cm-3 s-1 (σA) 6 × 10-9 cm-1) at 70 wt % under
nonvolcanic conditions is significantly higher than our estimate
for this σA (1.6 molecule cm-3 s-1), which may be due to their
overestimation of γ.
Our measurements indicate that kII can be higher (see Table
2) than the rate suggested by Burley and Johnston7 for this
reaction to become important in the stratosphere (kII g 6 × 102
M-1 s-1). From our estimated processing rates however, this
reaction is not significant even though kII ∼5 × 104 M-1 s-1 in
60 wt % acid at 219 K. The high solubility of HONO assumed
by Burley and Johnston is the main reason for their calculated
processing rates for HCl being high. The opposing variations
in HONO and HCl solubility with wt % sulfuric acid, which
was not anticipated by Burley and Johnston, is the main reason
this reaction is not important in the stratosphere.
Under volcanic conditions (for example, a log-normal dis-
tribution with rp ) 0.35 µm, log σ ) 0.2, and total number
density ) 35 particles cm-3; results in σA ) 8.4 × 10-7 cm-1),38
we estimate a ClNO formation rate of ∼1500 molecule cm-3
s-1. Zhang et al. estimated ∼5000 molecule cm-3 s-1 for
comparable σA; however, their estimation procedure did not take
into account any size dependence to γ. For d[HCl]/dt ) 1500
molecule cm-3 s-1, HCl processing by sulfuric acid particles is
comparable to that of reaction with OH. Note that this estimate
is based on a highly perturbed sulfuric acid aerosol layer and a
relatively high [HONO] of 1 × 107 molecule cm-3. We should
reiterate here that it is necessary to calculate γ under strato-
spheric conditions before making estimates about production
rates. Zhang et al. assumed γHONO ) 0.07, which is not
applicable for the reaction occurring in 70 wt % acid solutions
at stratospheric conditions.
Surface Reaction. Evidence for the reaction of HONO with
HCl occurring on the surface is: (i) γHONO is not proportional
to (pHCl)1/2 for many experiments and (ii) γHONO on HCl-doped
sulfuric acid aerosol has little or no dependence on particle size
(l e 0.04 µm). Hanson and Ravishankara,28 Hu et al.,39a and
Jayne et al.39b,c have previously identified surface contributions
in heterogeneous reactions. These observations are typically
Our measured rate coefficients in the 60 wt % solutions are
much higher than those in 50 wt %. Although it is not possible
to substantiate mechanisms using kinetic measurements, this is
consistent with the solubility measurements that suggest there
is a change in the N(III) species from 50 to 60 wt %. Apparently
the reaction of H2ONO+ with HCl/Cl- is faster than the reaction
of HONO with Cl-. kII in 70 wt % H2SO4 is higher yet at 215
K possibly indicating the presence of a third species, perhaps
NO+HSO4-. Another factor is the possible increase of kII with
acidity, as found for the reaction of HOCl with HCl in 49 to 67
wt % H2SO4.40 kII for HOCl + HCl increased about a factor
of 20 over this range; kII for HONO + HCl reported here
increases by about a factor of 100 from 50 wt % to 70 wt %
H2SO4 at 210 ( 5 K. The presence of a strong acid greatly
affects the solubility and reactivity of HONO in solution as
demonstrated by the null reaction in NaCl solution. This differs
from the suggestion of Fenter et al.11 where they suggested that
ClNO could be formed from reaction of HONO with solvated
chloride ions.
Conclusions
The effectiveness of this reaction under stratospheric condi-
tions is limited primarily by the fact that HONO and HCl are

3332 J. Phys. Chem. A, Vol. 102, No. 19, 1998
Longfellow et al.
(21) Patrick, R.; Golden, D. M. Int. J. Chem. Kinet. 1983, 15, 1189-
not mutually highly soluble in sulfuric acid. Over the range of
sulfuric acid content investigated here, the solubility of HONO
increases with increasing wt % H2SO4, whereas the HCl
solubility decreases as the acidity increases. Likewise, while
HCl becomes more soluble in dilute sulfuric acid, HONO
becomes less soluble. In addition, the relatively low reaction
rate coefficients (less than the diffusion-limited value of ∼109
1227.
(22) Huey, L. G.; Hanson, D. R.; Howard., C. J. J. Phys. Chem. 1995,
99, 5001-5008.
(23) Streit, G. E. J. Chem. Phys. 1982, 77, 826-833.
(24) Hanson, D. R.; Lovejoy, E. R. Geophys. Res. Lett. 1994, 22, 2401-
2404.
(25) Lovejoy, E. R.; Hanson, D. R. J. Phys. Chem. 1995, 99, 2080-
2087.
M-1 s-1 39
) obtained from the measurements here, lead to a very
(26) Huthwelker, T.; Peter, T.; Luo, B. P.; Clegg, S. L.; Carslaw, K.;
Brimblecombe, P. J. Atmos. Chem. 1995, 21, 81-95.
limited effect of this reaction, and it can probably be ignored
in the unperturbed stratosphere. Under certain volcanic condi-
tions, the processing rate of HCl by the heterogeneous reaction
with HONO can be comparable to that of reaction with OH,
but it will require a better estimate of stratospheric [HONO]
and detailed modeling calculations to determine its possible
impact.
(27) It is possible to predict kr for a reaction that is first order in both
directions if kf and K are known. (Atkins, P. W. Physical Chemistry, 3rd
ed.; Freeman: New York, 1986; p 702.) However, in uptake experiments,
the observed kf and kr do not necessarily reflect the kf and kr in solution.
The observed kf and kr depend on the time-dependent uptake into solution
and it is not possible to treat the data in the simple manner whereupon the
asymptotic value is subtracted from the signal and the resulting first-order
loss rate is set equal to kf + kr. The equation we used to estimate the amount
of HONO to subtract from the signal at injector position z is
In summary, the heterogeneous reaction of HONO with HCl
is apparently enhanced at the surface (with respect to the bulk
rate), and we observed this in both wetted wall flow tube
measurements and measurements on particles. For the condi-
tions of our experiments, it appears that the surface specific
component is expressed most at temperatures of 250 K and
above; the data at the lowest temperatures can be described using
a very small or nonexistent surface component.
z
Vave
HONOz
∆SHONO
)
kClNOSClNO3.6 exp-k
where Vave is the average carrier flow velocity, kClNO is the first-order loss
rate for ClNO obtained from ClNO uptake experiments, 3.6 is the sensitivity
factor between the signals due to ClNO and HONO (SClNO and SHONO
,
respectively), and kHONO (cm-1) is the first-order loss rate for HONO we
are seeking. kHONO was then obtained by an iterative procedure. In our
analysis, z was measured from a reference point that was ∼2 cm upstream
of the end of the liquid. This equation is a crude way to take into account
the reverse reaction. Note that the ∆SHONO values calculated from this
equation were comparable to those calculated from the equilibrium constant
and the estimated asymptotic signal levels.
Acknowledgment. This research was funded in part by
NOAA’s Climate and Global Change research program.
(28) Hanson, D. R.; Ravishankara, A. R. J. Phys. Chem. 1994, 98, 5728-
5735.
References and Notes
(1) Solomon, S. ReV. Geophys. 1988, 26, 131-148.
(2) Wofsy, S. C.; Molina, M. J.; Salawitch, R. J.; Fox, L. E.; McElroy,
M. B. J. Geophys. Res. 1988, 93, 2442-2450.
(29) Hanson, D. R. J. Phys. Chem. 1997, 101B, 4998-5001.
(30) Luo, B. P.; Clegg, S. L.; Peter, T.; Muller, R.; Crutzen, P. J.
Geophys. Res. Lett. 1994, 21, 49-52.
(3) For example, see: reports from the Airborne Antarctic Ozone
Expedition. J. Geophys. Res. 1989, 94, 11179-11738 and 16437-16860.
(4) Kolb, C. E.; Worsnop, D. R.; Zahniser, M. S.; Davidovits, P.;
Keyser, L. F.; Leu, M.-T.; Molina, M. J.; Hanson, D. R.; Ravishankara, A.
R.; Williams, L. R.; Tolbert, M. A. In Progress and Problems in
Atmospheric Chemistry; World Scientific: Singapore, 1995; Vol. 3.
(5) Steele, H. M.; Hamil, P. J. Aerosol Sci. 1981, 12, 517-528.
(6) Hanson, D. R.; Ravishankara, A. R.; Solomon, S. J. Geophys. Res.
1994, 99, 3615-3629.
(31) Vogt, R.; Finlayson-Pitts, B. J. Geophys. Res. Lett. 1994, 21, 2291-
2294.
(32) Giauque, W. F.; Hornung, E. W.; Kunzler, J. E.; Rubin, T. R. J.
Am. Chem. Soc. 1960, 82, 62-70. Gmitro, J. I.; Vermeulen, T. A. I. Chem.
Eng. J. 1964, 10, 740-746.
(33) Williams, L. R.; Golden, D. M. Geophys. Res. Lett. 1993, 20, 2227-
2230.
(34) There appear to be two contradictory spectroscopic measurements
of HONO and its protonated forms in the literature. One paper (Seel, F.;
Winkler, R. Z. Phys. Chem. NF 1960, 25, 217-232) apparently reports
equal amounts of HONO and NO+ at 56 wt % while another (Deschamps,
J. M. R. Comput. Rend. 1957, 245, 1432-1434) identifies H2ONO+ at 70.5
wt %, equal amounts of H2ONO+ and NO+ at 88 wt %, and only NO+ at
96 wt %.
(7) Burley, J. D.; Johnston, H. S. Geophys. Res. Lett. 1992, 19, 1363-
1266.
(8) JANAF Thermochemical Tables, J. Phys. Chem. Ref. Data 1985,
14 (Suppl. 1). Note the entropy for ClNO reported in ref 7 is wrong.
(9) Zhang, R.; Leu, M. T.; Keyser, L. F. J. Phys. Chem. 1996, 100,
339-345.
(10) Becker, K. H.; Kleffmann, J.; Kurtenbach, R.; Wiesen, P. J. Phys.
Chem. 1996, 100, 14984-14990.
(11) Fenter, F. F.; Rossi, M. J. J. Phys. Chem. 1996, 100, 13765-13775.
(12) Schwartz, S. E. Chemistry of Mulitphase Atmospheric Systems;
NATO ASI Series; Springer-Verlag: Berlin, 1986; pp 415-471.
(13) Hanson, D. R.; Ravishankara, A. R. J. Phys. Chem. 1993, 97,
12309-12319.
(14) Lovejoy, E. R.; Huey, L. G.; Hanson, D. R. J. Geophys. Res. 1995,
100, 18775-18780.
(15) Hanson, D. R.; Lovejoy, E. R. J. Phys. Chem. 1996, 100, 6397-
6405.
(16) Carslaw, K. S.; Clegg, S. L.; Brimblecombe, P. J. Phys. Chem.
1995, 99, 11557-11574.
(35) Arnold, F.; Scheid, J.; Stilp, Th.; Schlager, H.; Reinhardt, M. E.
Geophys. Res. Lett. 1992, 12, 2421-2424.
(36) Hanson, D. R.; Ravishankara, A. R.; Lovejoy, E. R. J. Geophys.
Res. 1996, 101, 9063-9069.
(37) For the reaction of OH + HCl a rate coefficient of 5.8 × 10-13
cm3 molecule-1s-1 at 212 K was used. Battin-Leclerc, F.; Kim, I. K.;
Talukdar, R. K.; Ravishankara, A. R. Manuscript in preparation.
(38) Deshler, T.; Hofmann, D. J.; Johnson, B. J.; Rozier, W. R. Geophys.
Res. Lett. 1992, 19, 199-202.
(39) (a) Hu, J. H.; Shi, Q.; Davidovits, P.; Worsnop, D. R.; Zahniser,
M. S.; Kolb, C. E. J. Phys. Chem. 1995, 99, 8768-8775. (b) Jayne, J. T.;
Duan, S. X.; Davidovits, P.; Worsnop, D. R.; Zahniser, M. S.; Kolb, C. E.
J. Phys. Chem. 1992, 96, 5452-5460. (c) Jayne, J. T.; Davidovits, P.;
Worsnop, D. R.; Zahniser, M. S.; Kolb, C. E. J. Phys. Chem. 1990, 94,
6041-6048.
(40) Donaldson, D. J.; Ravishankara, A. R.; Hanson, D. R. J. Phys.
Chem. 1997, 101A, 4717-4725.
(41) Atkins, P. W Physical Chemistry, 3rd Ed.; Freeman: New York,
1986; p 741.
(17) Howard, C. J., J. Phys. Chem. 1979, 83, 3-9.
(18) Brown, R. L. J. Res. Natl. Bur. Stand. 1978, 83, 1-8.
(19) Hirschfelder, J. O.; Brid, R. B.; Spotz, E. L. Trans. ASME 1949,
921-937.
(20) Monchick, L.; Mason, E. A. J. Chem. Phys. 1961, 35, 1676-1697.