In situ monitoring of the NaCl + HNO3 surface reaction: The observation of mobile surface strings

Article abstract of DOI:10.1021/jp982910x

The reaction of single-crystal NaCl(100) with dry HNO₃ was monitored using atomic force microscopy in contact mode. Initial exposure to dry HNO₃ results in the growth of a two-dimensional metastable layer of NaNO₃ to nearly complete surface coverage. Exposure of this metastable NaNO₃ layer to small amounts of H₂O produces deliquescence of the surface with concomitant rearrangement of the NaNO₃ adlayer to form unusual mobile strings presumed to contain NaNO₃. These strings are transient and eventually crystallize to form rhombohedral NaNO₃ crystals sitting on top of the NaCl surface.

Full text of DOI:10.1021/jp982910x

8950
J. Phys. Chem. B 1998, 102, 8950-8953
In Situ Monitoring of the NaCl + HNO3 Surface Reaction: The Observation of Mobile
Surface Strings
Christopher D. Zangmeister and Jeanne E. Pemberton*
Department of Chemistry, UniVersity of Arizona, Tucson, Arizona 85721
ReceiVed: July 8, 1998; In Final Form: August 24, 1998
The reaction of single-crystal NaCl(100) with dry HNO
contact mode. Initial exposure to dry HNO results in the growth of a two-dimensional metastable layer of
NaNO to nearly complete surface coverage. Exposure of this metastable NaNO layer to small amounts of
O produces deliquescence of the surface with concomitant rearrangement of the NaNO adlayer to form
unusual mobile “strings” presumed to contain NaNO . These strings are transient and eventually crystallize
to form rhombohedral NaNO crystals sitting on top of the NaCl surface.
3
was monitored using atomic force microscopy in
3
3
3
H
2
3
3
3
Reactions of particulate NaCl formed from evaporated sea
salt aerosols with nitrogen oxides have received considerable
attention in recent years owing to potential terrestrial atmo-
spheric implications for the formation of reactive chlorinated
contact AFM is not destructive to either NaCl or NaNO3
13-15
1
surfaces.
Calibrated aliquots of dry HNO3 headspace over
a HNO3/H2SO4 solution or saturated H2O headspace over liquid
H2O were delivered via syringe injections to a glass gastight
AFM cell (Digital Instruments, model FC) containing a NaCl-
(100) crystal (International Crystal Laboratories.) On the basis
of the geometric parameters of the gastight AFM cell, the cell
1
-10
species.
These NaCl particles undergo heterogeneous
-
reactions with trace gases to form products that are Cl
deficient.11 The formation of gaseous HCl in the terrestrial
atmosphere has been hypothesized to be linked to such
-
2
3
-6
volume is estimated to be ca. 8 × 10 cm (3 × 10 mol at
STP). Images were collected in situ as the reaction proceeded
using a 2 Hz scan rate to ensure high image quality and fast
acquisition times (4 min/image.) The average RH of the
ambient air during these experiments was ca. 14 ( 3%. Thus,
the NaCl(100) surfaces were not yet covered by a full monolayer
1
2
reactions.
One such reaction is that of vapor-phase HNO3 with NaCl
as shown below:
HNO (g) + NaCl(s) f NaNO (s) + HCl(g)
(1)
3
3
of adsorbed H2O, since it has been shown that, at RH below
The mechanism and kinetics of reaction 1 have been extensively
16
3
0%, H2O adsorbs as two-dimensional clusters.
studied,1
,4,7,8,10
and it has been shown that the reaction prob-
Prior to analysis, a 1 cm × 1 cm single-crystal NaCl(100)
ability is increased by the presence of surface defects and
was freshly cleaved in ambient air and mounted in the gastight
cell. Aliquots of dry HNO3 and H2O vapor were dosed onto
the NaCl surface as appropriate. Although very small amounts
of NO2 and H2O might be present in the dosed vapor from
dissociated HNO3 vapor, the amounts of these species, if present,
1
,4,7-10
adsorbed H2O.
H2O adlayers allow dissolution of the
surface to form a saturated NaCl solution (surface deliquescence)
into which HNO3 can diffuse and react.
Previous work has shown that upon exposure of NaCl to
HNO3, a layer of 1-2 monolayers of NaNO3 caps the surface,
thereby protecting it from further reaction.8 Upon addition of
H2O, the surface rearranges to form crystals of NaNO3 sitting
on top of the NaCl, exposing fresh NaCl that reacts further with
7
are too low to detect by mass spectrometry and are therefore
,9
assumed to be negligible. It is further assumed that the uptake
of dosed species by the AFM cell walls is negligible. Image
acquisition was initiated immediately after dosing.
1
,4,7-9
HNO3.
X-ray photoelectron spectroscopy (XPS) results
Three 5 µm × 5 µm AFM images collected prior to and after
dry HNO3 dosing are shown in Figure 1. Figure 1a shows a
contact AFM image of freshly cleaved NaCl(100) prior to HNO3
dosing. Monotomic steps (0.56 nm) are observed across the
indicate that upon exposure of this surface to H2O at relative
humidities (RH) well below the deliquescence point of NaCl,
ca. 80% of the surface becomes fresh NaCl after this NaNO3
8
,9
rearrangement.
The remaining 20% of the surface is com-
-
5
8
surface. After the addition of 1 mL of dry HNO3 (4 × 10
prised of rhombohedral NaNO3 crystals as shown by transmis-
3
sion electron microscopy.9
mol/cm ), a metastable NaNO3 layer forms as shown in Figure
1
b. For a dose of this volume, a NaNO3 layer caps the NaCl
The reaction of single-crystal NaCl(100) with dry HNO3
vapor using contact atomic force microscopy (AFM) is reported
here with the goal of better elucidating the mechanism of NaNO3
crystal growth. This reaction is attractive for study, because
its reaction kinetics are compatible with AFM imaging times
and control of HNO3 and H2O exposure is readily achieved.
A Digital Instruments Nanoscope IIIa AFM in contact mode
was used for these experiments. Previous work has shown that
within 4 min producing a surface with several small (ca. 10
nm diameter) chimney-like defects. These structures are
hypothesized to be small NaNO3 crystals that appear owing to
the presence of small amounts of H2O adsorbed on the NaCl
surface. XPS on a VG ESCALAB MKII using Al KR excitation
confirms the presence of a NaNO3 capping layer at this stage
of the reaction sequence. The N 1s and O 1s binding energies
are observed at 407.2 and 532.9 eV, respectively, indicative of
NaNO3. Subsequent HNO3 additions result in no further
morphological changes after the capping layer is established.
*
To whom correspondence should be addressed. Email: pembertn@
u.arizona.edu.
1
0.1021/jp982910x CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/16/1998

Letters
J. Phys. Chem. B, Vol. 102, No. 45, 1998 8951
-
5
3
Figure 1. Contact AFM images of (a) NaCl(100) surface prior to HNO
3
exposure; (b) surface in (a) exposed to 4 × 10 mol/cm HNO
3
; (c)
-
5
3
subsequent exposure of surface in (b) to 1 × 10 mol/cm H
2
O after (a).
TABLE 1: Roughness Analysis Calculations on NaCl(100) as a Function of Exposure
reactive exposure
-
5
-5
-4
-4
4
× 10 mol/cm3 9 × 10 mol/cm3 1 × 10 mol/cm3 1 × 10 mol/cm3
× 10 mol/cm3 HNO
-
5
+ 1 × 10
-5
HNO
+ 1 × 10
-5
HNO
+ 4 × 10
-5
HNO
3
+ 1 × 10
-4
4
3
3
3
3
3
3
3
HNO
3
mol/cm H
2
O
mol/cm H
2
O
mol/cm H
2
O
mol/cm H
2
O
total surface area (µm2)
surface area above 2 nm (µm )
surface area above 2 nm
maximum roughness feature height (nm)
25.0
0.3
1.2
32
26.3
6.0
22.7
136
25.5
6.6
26.0
119
26.9
9.0
33.3
203
27.3
12.3
45.2
244
2
%
To study the effect of H2O exposure on the NaNO3 capping
surface roughens with HNO3 exposure, and the growth of what
is presumed to be NaNO3 occurs preferentially along step edges.
In addition, small spindles of NaNO3 of ca. 1.2 nm in height
(ca. 4 monolayers) grow on terrace sites parallel to the step
edges. The surface continues to roughen after 24 min exposure
-
5
3
layer, a 1 mL (1 × 10 mol/cm ) H2O-saturated headspace
vapor aliquot was added to the cell resulting in the image shown
in Figure 1c. This volume exceeds the cell volume by a factor
of ca. 15; therefore, the entire cell volume remains saturated
with H2O vapor throughout the course of these experiments.
Upon H2O addition, the capping NaNO3 layer rearranges to form
well-defined NaNO3 “towers” with an average height of ca. 150
nm. These towers are ca. 15 times higher than any of the surface
features prior to H2O exposure (note the z-axis scale difference
between parts b and c of Figure 1). The regions between the
NaNO3 towers appear to be NaCl(100) based on a maximum
feature height of ca. 1 nm.
Quantitative determination of the relative proportions of the
NaNO3 crystal and unreacted NaCl surface areas from these
images was undertaken by conservatively assuming that any
area above 2 nm was NaNO3. The results of this analysis are
shown in Table 1. Prior to H2O exposure, only ca. 1% of the
-
7
3
to this single aliquot of 4 × 10 mol/cm HNO3 vapor (Figure
2c). It is important to note that, at this stage of the reaction,
the NaCl(100) surface morphology is generally maintained as
the reaction proceeds; surface deliquescence due to adsorbed
or dosed H2O has not yet occurred. To achieve nearly complete
coverage by NaNO3 (Figure 2d), a total of 40 µL of HNO3 vapor
-
6
3
(1 × 10 mol/cm ) were injected. A cross-sectional analysis
of the image in Figure 2d (not shown) indicates a smoother
and thinner NaNO3 layer capping the underlying NaCl substrate
than observed for larger HNO3 exposures (Figure 1b), and NaCl
step edges are still evident.
-5
3
Exposure of this surface to 1 mL (1 × 10 mol/cm ) of H2O-
saturated headspace results in the images shown in Figure 3.
This H2O exposure is sufficient to deliquesce the underlying
NaCl surface. The NaCl(100) surface morphology evident in
Figure 2a-d is replaced by a collection of unusual “strings”
presumed to contain NaNO3 unevenly distributed over the NaCl
surface. Interestingly, these strings do not follow the previously
observed step edges of the NaCl(100) surface (Figure 3a).
A magnified 1 µm × 1 µm area of Figure 3a is shown in
Figure 3b. The NaNO3-containing strings are clearly evident
in this presentation and are observed independent of the scan
direction. Small (<0.2 nm) roughness features evident in Figure
3b that run parallel to the scan direction are attributed to tip-
induced defects. A cross-sectional analysis of this image shown
in Figure 4 reveals these strings to be of remarkably consistent
height (ca. 1.2 nm) and thickness (ca. 80 nm).
-
5
surface is above 2 nm in height. After exposure to 1 × 10
3
mol/cm of H2O, ca. 23% of the surface is composed of NaNO3
crystals. This value is in excellent agreement with previous
XPS results, which indicate ca. 20% of the surface as NaNO3
crystals after H2O exposure.8
Additional doses of H2O and HNO3 after this NaNO3
crystallization (Figure 1c) result in an increase in coverage and
average height of the NaNO3 crystals. This behavior reflects
continued reaction of HNO3 with the exposed NaCl to form
crystalline NaNO3 in the presence of adsorbed H2O. Thus, the
presence of a H2O adlayer facilitates NaNO3 formation.
Presumably, NaNO3 formation occurs as a precipitation reaction
4
,6
from saturated solution as opposed to solid-state reaction.
A second set of contact AFM experiments was performed to
monitor the growth of NaNO3 under lower HNO3 exposure (ca.
00 times lower exposure than the previous experiments.)
The number of strings and their positions across the NaCl
surface change with time. The image in Figure 3c was acquired
4 min after the image in Figure 3a at the exact same spot on
the surface. Despite the obvious mobility of these strings, their
average height and thickness are maintained. Notable in both
images are strings that are very close together yet that do not
lead to NaNO3 crystallization.
1
Figure 2a shows a 5 µm × 5 µm image of a freshly cleaved
NaCl(100) surface. Prior to HNO3 exposure, the surface has
numerous monatomic steps running ca. 45° to the scan direction.
-
7
Figure 2b was acquired 2 min after a 10 µL aliquot (4 × 10
3
mol/cm ) of dry HNO3 was injected into the gastight cell. The

8952 J. Phys. Chem. B, Vol. 102, No. 45, 1998
Letters
exposure; (b) subsequent exposure of surface in (a) to 4 × 10 mol/cm3
-7
Figure 2. Contact AFM images of (a) NaCl(100) surface prior to HNO
dry HNO
surface in (a) to 1 × 10 mol/cm HNO
3
-
7
3
3
; (c) 24 min after image in (b); (d) subsequent exposure of surface in (c) to an additional 6 × 10 mol/cm HNO
3
(total exposure of
-
6
3
3
).
-
6
3
-5
3
Figure 3. Contact AFM of NaCl(100) surface exposed to 1 × 10 mol/cm HNO
µm image of NaNO
strings on NaCl; (b) 1 µm × 1 µm zoom of boxed region in image (a); (c) 5 µm × 5 µm image of strings 4 min after image
in (a); (d) 5 µm × 5 µm image after NaNO crystallization.
3
followed by exposure to 1 × 10 mol/cm H O. (a) 5 µm ×
2
5
3
3
NaNO3 crystallization at this spot on the surface is evident
in the image acquired ca. 8 min after this single H2O exposure.
Figure 3d shows one such crystal of ca. 30 nm in height that
has apparently formed from the upper area in Figure 3c in which
the strings have formed a completely closed pseudo-rectangular
arrangement. Significantly, no strings are evident on the surface
in the immediate area surrounding the crystal; however, sub-
nanometer string “tracks” are observed on the NaCl at positions

Letters
J. Phys. Chem. B, Vol. 102, No. 45, 1998 8953
3
Figure 4. Cross-sectional analysis of NaNO strings across the NaCl(100) surface.
where the strings previously existed. At the end of this surface
reaction induced by exposure to H2O, the surface is covered by
ca. 30 µm rhombohedral NaNO3 crystals larger than those shown
in Figure 1c.
Acknowledgment. The authors gratefully acknowledge
support of this research by the National Science Foundation
(CHE-9504345.) The authors thank Dr. Andrew Back for
computational assistance and useful discussions on the NaNO3/
NaCl interface.
The origin of the strings observed after surface deliquescence
is rationalized as follows. Exposure of the NaNO3-capped
surface to H2O results in displacement of the metastable NaNO3
layer with concomitant deliquescence of the underlying NaCl
surface, the energetically most favorable process. (The ener-
getics of this system are probably dictated by the lack of a
commensurate lattice between NaNO3 and NaCl as ascertained
References and Notes
(1) Vogt, R.; Finlayson-Pitts, B. J. J. Phys. Chem. 1994, 98, 3747.
(
2) Timonen, R. S.;, Chu, L. T.; Leu, M.; Keyser, L. F. J. Phys. Chem.
1
994, 98, 9509.
17
by computer calculations using Epicalc. ) The formation of
strings containing NaNO3 that are mobile is attributed to the
presence of a H2O adlayer on the NaCl surface and the lack of
a strong interaction between NaNO3 and the NaCl surface. The
transient nature of these strings suggests a kinetic barrier
associated with the crystallization of NaNO3. If NaNO3
crystallization occurs as a precipitation reaction from saturated
aqueous solution, then this kinetic barrier could be the dissolu-
tion of these strings in the underlying H2O adlayer.
(
(
3) Karlsson, R.; Ljungstr o¨ m, E. J. Aerosol. Sci. 1995, 26, 39.
4) Leu, M.; Timonen, R. S.; Keyser, L. F.; Yung, Y. Y. J. Phys. Chem.
1995, 99, 13203.
(5) Caloz, F.; Fenter, F. F.; Rossi, M. J. J. Phys. Chem. 1996, 100,
494.
7
(
(
(
6) Peters, S. J.; Ewing, G. E. J. Phys. Chem, 1996, 100, 14093.
7) Beichert, P.; Finlayson-Pitts, B. J. J. Phys. Chem. 1996, 100, 15228.
8) Laux, J. M.; Fister, T. F.; Finlayson-Pitts, B. J.; Hemminger, J. C.
J. Phys. Chem. 1996, 100, 19891.
(9) Allen, H. C.; Laux, A. J. M.; Vogt, R.; Finlayson-Pitts, B. J.;
Hemminger, J. C. J. Phys. Chem. 1996, 100, 6371.
10) Langer, S.; Pemberton, R. S.; Finlayson-Pitts, B. J. J. Phys. Chem.
A 1997, 101, 1277.
11) Hitchcock, D. R.; Spiller, L. L.; Wilsson, W. E. Atmos. EnViron.
1980, 14, 165.
(12) Keene, W. C.; Pszenny, A. A. P.; Jacob, D. J.; Duce, R. A.;
Galloway, J. N.; Schultz-Tokos, J. J.; Sievering, H.; Boatman, J. F. Global
Biochem. Cycles 1990, 4, 407.
13) Shindo, H.; Ohashi, M.; Tateishi, O.; Seo, A. J. Chem. Soc.,
Faraday Trans. 1997, 93, 1169.
(14) Shindo, H.; Ohashi, M.; Baba, K.; Seo, A. Surf. Sci. 1996, 357-
58, 111.
15) Nakahara, S.; Langford, S. C.; Dickinson, J. T. Tribol. Lett. 1995,
, 277.
16) Peters, S. J.; Ewing, G. E. Langmuir 1997, 13, 6345.
17) Last, J. A.; Hiller, A. C.; Hooks, D. E.; Maxson, J. B.; Ward, M.
In summary, contact AFM has been used to observe in situ
the reaction of HNO3 with NaCl. Exposures of NaCl(100) to
(
-
5
3
large amounts of HNO3 (4 × 10 mol/cm ) with subsequent
exposure to H2O results in a surface containing rhombohedral
NaNO3 crystals (ca. 23% of the surface area) with the remaining
surface as unreacted NaCl. The newly exposed NaCl surface
further reacts with HNO3 to form additional NaNO3. Images
(
(
-
7
3
acquired after low HNO3 exposure (4 × 10 mol/cm ) show
NaNO3 growth along NaCl step edges. The NaNO3 layer con-
tinues to grow two-dimensionally until complete surface cover-
age is achieved. Subsequent H2O exposure results in surface
deliquescence and the formation of mobile strings containing
NaNO3. These strings are transient in nature with the formation
of rhombohedral NaNO3 crystals as the final surface state.
3
1
(
(
(
D. Chem. Mater. 1998, 10, 422.