Properties of Crystalline Sulfuric Acid in Ultrahigh Vacuum: Incongruent Melting to the Liquid Azeotrope and Its Slow Recrystallization

Article abstract of DOI:10.1021/jp962251t

The growth and characterization of ultrathin (ca. 80 monolayers) films of pure, crystalline sulfuric acid and of the liquid H2SO4 + H2O azeotrope in ultrahigh vacuum (UHV) are described.The films, which were deposited on a Pt(111) substrate, were synthesized in vacuo via the reaction of co-condensed mixtures of H2O and SO3.Fourier transform infrared reflection-absorption spectroscopy (FTIRAS) shows that the pure, crystalline acid forms when such mixtures are briefly annealed at 260 K.Crystalline H2SO4 is stable in UHV until 285 K, at which temperature FTIRAS measurements suggest the melting occurs.The melting temperature is equal, within experimental error, to the melting point of a macroscopic sample of crystalline H2SO4.Analysis of the gas phase composition during melting reveals that SO3 is expelled from a film as it melts.The liquid is identified as the azeotrope of H2SO4 and H2O, formed in the incongruent melting of H2SO4.The H2SO4 + H2O mixtures can be supercooled to 255 K.Freezing results in the formation of pure, crystalline H2SO4 and is accompanied by the expulsion of H2O into the gas phase.To our knowledge, this work provides the first observation of melting of a molecular solid under UHV.

Full text of DOI:10.1021/jp962251t

19582
J. Phys. Chem. 1996, 100, 19582-19586
Properties of Crystalline Sulfuric Acid in Ultrahigh Vacuum: Incongruent Melting to the
Liquid Azeotrope and Its Slow Recrystallization
Liesl R. Schindler and Jeffrey T. Roberts*
Department of Chemistry, UniVersity of Minnesota, Minneapolis, Minnesota 55455-0431
ReceiVed: July 24, 1996X
The growth and characterization of ultrathin (≈80 monolayers) films of pure, crystalline sulfuric acid and of
the liquid H2SO4 + H2O azeotrope in ultrahigh vacuum (UHV) are described. The films, which were deposited
on a Pt(111) substrate, were synthesized in Vacuo via the reaction of co-condensed mixtures of H2O and SO3.
Fourier transform infrared reflection-absorption spectroscopy (FTIRAS) shows that the pure, crystalline acid
forms when such mixtures are briefly annealed at 260 K. Crystalline H2SO4 is stable in UHV until 285 K,
at which temperature FTIRAS measurements suggest the melting occurs. The melting temperature is equal,
within experimental error, to the melting point of a macroscopic sample of crystalline H2SO4. Analysis of
the gas phase composition during melting reveals that SO3 is expelled from a film as it melts. The liquid is
identified as the azeotrope of H2SO4 and H2O, formed in the incongruent melting of H2SO4. The H2SO4 +
H2O mixtures can be supercooled to 255 K. Freezing results in the formation of pure, crystalline H2SO4 and
is accompanied by the expulsion of H2O into the gas phase. To our knowledge, this work provides the first
observation of melting of a molecular solid under UHV.
Introduction
H2SO4 + H2O azeotrope, and we consider the kinetics and
thermodynamics of the solid-liquid and liquid-solid phase
transitions.
The surface chemical properties of sulfuric acid are of
relevance to a wide range of environmentally and technologi-
cally important phenomena, such as heterogeneous atmospheric
chemistry, electrochemistry, and corrosion.1-3 Critical to an
understanding of sulfuric acid surface chemistry is a corre-
sponding understanding of adsorbate structure and adsorption/
desorption kinetics. Few methods are available to study these
issues, partly because many conveniently implemented analytical
probes (infrared spectroscopy, mass spectrometry, etc.) are
intrinsically bulk sensitive, preventing their application to
macroscopic sulfuric acid samples, which have small surface
area-to-volume ratios. By studying sulfuric acid films between
10 and 100 monolayers thick, it should be possible to circumvent
surface area-to-volume ratio limitations. Surface analysis could
be further simplified by conducting experiments in ultrahigh
vacuum (UHV), where numerous surface-sensitive spectroscopic
probes are available. It has been demonstrated that this
approachsthe study ultrathin films in UHVscan be used to
study the surface chemical properties of ice.4-6 The approach
has recently been extended to interfacial systems involving
sulfuric acid.7
Experimental Section
Experiments were conducted in a UHV chamber that is
described in detail elsewhere.9 Sulfuric acid films were
deposited on a 111-oriented, single crystal platinum substrate
that was mounted on a sample manipulator capable of x-, y-,
and z-translation and of rotation about the z-axis. The substrate
was in thermal contact with a liquid nitrogen-cooled reservoir
and could be cooled to approximately 90 K. Radiative heating
was provided by a tungsten filament ≈2 mm behind the crystal.
Temperatures were measured with a chromel/alumel thermo-
couple junction that was spot-welded to the Pt substrate; an
electronic ice point substituted for a reference junction. Gases
were admitted into the chamber using directed dosers.
The UHV chamber is equipped with instrumentation for
Fourier transform reflection-absorption infrared spectroscopy
(FTIRAS) and temperature-programmed desorption (TPD). For
FTIRAS, the infrared beam was reflected from the Pt substrate
at an angle of 4° from the surface plane. Desorption data were
acquired with the substrate positioned in line of sight of a mass
spectrometer that was interfaced to a data acquisition board.
As many as seven ion-temperature profiles could be collected
during a single experiment.
For sulfuric acid, the experimental challenge of a UHV-based
approach is to devise a method for film deposition in UHV.
The vapor pressure of pure sulfuric acid (≈10-3 Pa at room
temperature)8 is too low for a method based on direct condensa-
tion to be practicable, particularly since sulfuric acid is
exceedingly corrosive. Earlier, we described a method for the
synthesis of ultrathin sulfuric acid films in Vacuo.7 The method
is very flexible, allowing the synthesis of pure sulfuric acid as
an amorphous or crystalline solid. Mixtures of sulfuric acid
and water can be deposited as well, and their compositions are
variable over a wide range. In this paper we further explore
the properties of sulfuric acid in UHV. In particular, we show
that pure, crystalline sulfuric acid reversibly melts to form an
The synthesis of ultrathin H2O + H2SO4 films is described
elsewhere.7 Briefly, H2O/SO3/H2O “sandwiches” are deposited
on the Pt substrate at 90 K. The sandwiches are converted to
sulfuric acid by briefly annealing at a temperature between 180
and 260 K, depending on the desired film composition. Film
thicknesses cited in this work are given in units of monolayer
equivalents (ML) and are based upon the H2SO4 desorption yield
during TPD. The water used to synthesize sulfuric acid was
deionized and triply distilled and was degassed via several
freeze-pump-thaw cycles before use each day. Water-d2 (CIL)
and sulfur trioxide (Aldrich) were degassed before use each day
and otherwise used as received.
* To whom correspondence should be addressed. Telephone (612) 625-
2363, fax (612) 626-7541, e-mail roberts@chem.umn.edu.
X Abstract published in AdVance ACS Abstracts, November 15, 1996.
S0022-3654(96)02251-4 CCC: $12.00 © 1996 American Chemical Society

Crystalline Sulfuric Acid in Ultrahigh Vacuum
J. Phys. Chem., Vol. 100, No. 50, 1996 19583
important to emphasize the spectral changes do not occur
gradually with increasing temperature. Rather, they are abrupt,
an observation that is strongly suggestive of a phase transition
from the solid to the liquid. Notably, the changes occur at a
temperature that is, within the uncertainty of the measurements,
equal to what Bolsatis reports as the melting point, Tm, of
crystalline sulfuric acid (283.45 K).8
A second important spectral change that occurs with increas-
ing temperature is that absorbance decreases above ≈270 K.
This is because the sulfuric acid evaporation rate becomes
significant (>0.1 ML s-1). The lifetime of sulfuric acid film
is obviously dependent on temperature and thickness. From
FTIRAS measurements, we estimate that an 80 ML evaporates
in approximately 2 min at 280 K.
Evaporation below and above the Melting Point. The
evaporation of sulfuric acid was studied with TPD, the results
of which provide further evidence for melting. The following
ions were monitored mass spectrometrically: m/e 18 (H2O+),
m/e 48 (SO+), m/e 64 (SO2+), m/e 80 (SO3+), and m/e 98 (H2-
SO4+). In Figure 2 are shown representative m/e 18, 80, and
98 ion traces from the evaporation of an 80 ML crystalline H2-
SO4 film at a heating rate of 0.5 K s-1. The spectra can be
divided into two temperature regions, below and above the H2-
SO4 melting point of 283.45 K. Below Tm, the line shapes of
the ion traces are essentially identical; the spectra differ only
in intensity. The mass spectrum of anhydrous sulfuric acid has
not to our knowledge been reported, but measurements of the
Figure 1. Single reflection FTIR spectra of (a) crystalline H2SO4 at
100 K, (b) liquid H2SO4 at 285 K, (c) supercooling H2SO4 at 273 ( 2
K, (d) crystallizing H2SO4 at 257 ( 2 K, and (e) recrystallized H2SO4
at 240 ( 2 K. Absorbance scales are not constant between spectra.
The two zeroth-order plots are offset in order to better discern the slope
change at Tm.
concentrated acid suggest that the H2SO4 parent ion fragments
+ 14
to H2O+, SO+, SO2+, and SO3
.
Also, although gaseous H2-
SO4 coexists with H2O and SO3 at equilibrium, H2SO4 dis-
sociation should be extremely slow in the nearly collisionless
UHV environment. The early leading edges of the spectra in
Figure 2 are thus attributed to H2SO4 vapor formed in the
sublimation of the solid film.
Results
Spectroscopic Observation of the Solid-Liquid Phase
Transition in UHV. The infrared spectrum of a pure sulfuric
acid film, prepared as described above with a 270 K annealing
temperature, is shown in Figure 1a. Because the infrared beam
is reflected from the underlying Pt(111) substrate, the spectrum
contains information about bulk and surface structure. Strategies
for the discernment of features associated with the film surface
are discussed elsewhere; here we are concerned only with
vibrational modes derived from the bulk.9,10 The spectrum in
Figure 1a, which was recorded at 100 K, can be unambiguously
assigned to the pure, crystalline acid. In particular, the sharp
band in the OH stretching region at 2980 cm-1 is consistent
only with crystalline H2SO4; spectra of amorphous and liquid
sulfuric acid, of H2O + H2SO4 solutions, and of the crystalline
sulfuric acid hydrates (H2SO4‚nH2O) exhibit much broader OH
stretching bands. None of the vibrational features are attribut-
able to H2S2O7, the anhydride of sulfuric acid.11 Assignments
of all the principal vibrational features are summarized in Table
1. The assignments are based upon those given by Gigue´re
and Savoie to the spectra of macroscopic H2SO4 samples.11,12
The assignments are supported by the vibrational frequencies
of ultrathin, crystalline D2SO4 films, which are also summarized
in Table 1.
The frequencies and relative intensities of the vibrational
modes of crystalline sulfuric acid do not change significantly
with temperature until 285 ( 2 K. At that temperature, the
spectra undergo changes that indicative of melting (Figure 1b
and Table 1). All of the bands, particularly that associated with
OH stretching, broaden, suggestive of a decrease of order in
the films. In addition, the S-O-H asymmetric bend becomes
much more intense, and it shifts in frequency from 1241 to 1291
cm-1. An analogous change has been reported when macro-
scopic samples of crystalline sulfuric acid are melted.13 It is
Above Tm, two aspects of the evaporation spectra are worthy
+
of note. First, at 285 K, the SO3 intensity begins to increase
more rapidly than do the H2O+ and H2SO4 intensities. The
+
rapid increase is brief (close to the data acquisition rate), and
by 287 K, the SO3+ signal again changes at approximately the
same rate as the H2O+ and H2SO4 signals. Similar changes
+
are observed in the SO+ and SO2+ spectra, but not in the H2O+
and H2SO4 spectra. The transient bursts of SO+, SO2+, and
+
SO3+ strongly imply that melting is incongruent: H2SO4 melts
to form a H2O + H2SO4 mixture, and excess SO3 is expelled
into the gas phase. A second noteworthy feature of the
desorption spectra is that the ion traces have different line shapes
above the melting point. This is a manifestation of the fact
that the liquid is a binary system, while the solid is a quasi-
single-component system. The vapor pressures of H2O and H2-
SO4 in a H2O + H2SO4 mixture do not vary with temperature
in the same way, and as a result, the composition of the liquid
changes with temperature.
The rate expression for sulfuric acid evaporation depends
upon temperature. Specifically, evaporation is roughly zeroth
order both above and below the melting point, but the evapora-
tion energies are different (inset of Figure 2).15 Below Tm, the
evaporation energy, which corresponds to the H2SO4 sublimation
energy (Esub), is 93 ( 5 kJ mol-1. Above Tm, the vaporization
energy (Evap) is 82 ( 3 kJ mol-1. These values are in close
agreement with those reported by Bolsatis for the sublimation
and vaporization of sulfuric acid.8 Their difference, 11 kJ
mol-1, is equal to the heat of fusion of H2SO4 reported by
Giauque.16 The trailing edges of the spectra drop of less rapidly
than is normally observed in the sublimation spectra of
condensed multilayers in UHV.15 This is attributed, at least in

19584 J. Phys. Chem., Vol. 100, No. 50, 1996
Schindler and Roberts
TABLE 1: Infrared Frequencies and Mode Assignments for H2SO4 and D2SO4
frequency/cm-1
crystalline H2SO4
(this work)
crystalline H2SO4
(ref 12)
crystalline D2SO4
(this work)
crystalline D2SO4
(ref 12)
liquid H2SO4
(this work)
liquid H2SO4
(ref 12)
mode
ν
ν
ν
ν
ν
ν
asym OH(D)
sym OH(D)
asym SdO
sym SdO
asym S-O-H
sym S-O-H
2980
2455
1395
1195
1241
1169
2970
2450
1365
1170
1240
1170
2260
1870
1370
1185
a
2280
1860
1350
1190
930
3000
2465
1400
1205
1291
1205 (sh)
2970
2450
1368
1195
nab
a
nab
1137
a Frequency below cutoff of infrared detector. b Not assigned.
Figure 2. Temperature-programmed desorption of a pure, solid sulfuric
acid film deposited on Pt(111). The heating rate was 0.5 K s-1. The
melting temperature is indicated by the arrow. The inset shows the
zeroth-order plots from which we determine a sublimation energy of
93 ( 5 kJ/mol and a vaporization energy of 82 ( 3 kJ/mol.
Figure 3. Temperature-programmed desorption of pure, solid sulfuric
acid film during heating to 290 K and subsequent cooling. See text
for discussion of important features. The heating rate was 0.5 K s-1
and the cooling rate was ≈2 K s-1
.
part, to pumping speed effects, since the initial layer was quite
thick (≈80 ML). It is also possible that the evaporation is not
zeroth order at high temperatures, due, for instance, to mor-
phological changes in the films.
respectively. (The temperature of a film decreased by ≈4 K
during the time required to obtain an infrared spectrum. The
temperatures cited above are those at the data acquisition
midpoints.) The spectra provide clear evidence for conversion
of the liquid H2SO4 + H2O mixture back to crystalline H2SO4
by 257 K. In particular, the broad OH stretching region narrows
and acquires a line shape essentially identical to that of the
originally crystalline film. The S-O-H asymmetric bond
changes as well, decreasing in intensity and shifting in frequency
back to 1241 cm-1. Interestingly, the spectra are attributable
to the liquid until ≈260 K, well below Tm.
The transient bursts of the SO+, SO2+, and SO3+ signals are
observed during TPD only when the heating rate (â) is e1 K
s-1. For â > 1 K s-1, the line shapes of the spectra differ above
285 K, but no pressure bursts are observed near Tm. TPD studies
of amorphous ice have shown that the crystallization and
sublimation rates are comparable at low â but that sublimation
dominates at high â.17,18 In the H2SO4 system, the disappear-
ance of desorption features associated with a phase transition
is associated with an instrumental artifact. The data acquisition
rate for each mass channel was ≈4 points per second. At
heating rates greater than 1 K s-1, adjacent data points are
separated by at least 0.25 K, over which temperature range
melting occurs. The gaseous products of incongruent melting
were therefore not observable at high â.
Mass spectrometric measurements give additional evidence
for recrystallization of the liquid sulfuric acid films upon
cooling, and they provide insight into the origin of hysteresis
in the liquid-solid phase transition. In Figure 3 are shown three
heating/cooling curves for a sulfuric acid film that was originally
crystalline and 80 ML thick. During heating (0.5 K s-1) and
cooling (2 K s-1), the mass spectrometer was tuned to m/e 98
(H2SO4+), m/e 80 (SO3+), and m/e 18 (H2O+). Important
features of the heating/cooling curves are as follows. First, mass
spectral intensity is always greater on the cooling part of the
curve. Second, there is no SO3+ burst during cooling. Finally,
Recrystallization of Liquid H2SO4 + H2O. The liquid
sulfuric acid films studied in this work could be recrystallized
under appropriate conditions. In Figure 1c-e are shown three
infrared spectra of a sulfuric acid film recorded as it cooled;
they were obtained at 273 ( 2, 257 ( 2, and 240 ( 2 K,

Crystalline Sulfuric Acid in Ultrahigh Vacuum
J. Phys. Chem., Vol. 100, No. 50, 1996 19585
water is released from a film as it cools through ≈255 K,
approximately the temperature at which infrared spectra indicate
that crystallization occurs. All of these phenomena are associ-
ated with the cooling and/or crystallization of liquid H2SO4 +
H2O. The release of water is a consequence of the fact that
freezing, like melting, is incongruent, since the composition of
crystalline H2SO4 is different from that of liquid H2SO4 + H2O.
Water evolution below Tm indicates that the liquids pass through
a metastable, supercooled state as they cool. That the m/e 98,
80, and 18 cooling curves are not superimposable on their
corresponding heating curves is at least partly attributable to
the fact that the cooling films are metastable between 285 and
255 K: the vapor pressure of a metastable substances is always
greater than that of a stable substance. Because the vacuum
system is subjected to a gas load during heating, the higher mass
spectrometer signals upon cooling may also be associated with
pumping speed effects. However, it is unlikely that pumping
speed effects account for the different heating and cooling curves
in Figure 3. In our hands, the heating and cooling curves of an
80 ML crystalline ice film are nearly superimposable. If
pumping speed effects were responsible for all hysteresis in the
sulfuric acid cooling curves, they should be manifest as well in
the cooling curves of ice.
were studied some time ago by Kunzler19 and Giauque.16
They found that the mole fraction of H2SO4 in an equilibrium
H2SO4 + H2O solution at its boiling point smoothly increases
with decreasing externally applied pressure, from 0.920 at 1.33
× 105 Pa to 0.937 at 1.33 × 104 Pa. Extrapolation to UHV,
where the pressure is essentially zero, implies an H2SO4 mole
fraction of between 0.94 and 0.95 in an ultrathin liquid film at
its melting point. Since vapor pressure increases with temper-
ature, the composition of an H2SO4 + H2O mixture in
equilibrium with the gas phase changes with temperature.
Specifically, the solutions gradually become enriched in water.
This accounts, at least in part, for the different line shapes of
the m/e 98, 80, and 18 TPD spectra above the sulfuric acid
melting point.
Mass spectrometric and FTIRAS measurements of the cooling
H2SO4 + H2O films show that they do not freeze at Tm, but
supercool, eventually crystallizing at approximately 257 K, ≈30
K below Tm. Supercooling is a well-known phenomenon in
concentrated sulfuric acid solutions, which are viscous and reach
equilibrium slowly.20 However, it is surprising that recrystal-
lization is inhibited to such an extent in a film only several
tens of monolayers thick. Slow recrystallization may be due
in part to the fact that freezing requires the migration of excess
H2O from the film interior to the surface and its subsequent
expulsion into the gas phase. In any case, because supercooled
films of ultrathin sulfuric acid films are so stable, UHV
experiments on the surface of liquid sulfuric acid are now
possible. At 262 K, just above the transition temperature from
the supercooled liquid to the crystalline solid, the vapor pressure
of sulfuric acid is so low that the lifetime of an 80 ML thick
film is several minutes. Under such conditions, reactive and
nonreactive sticking of numerous gases can be investigated, as
can adsorbate and absorbate structure. These subjects are
currently under investigation in this laboratory.
Preliminary work suggests that the sulfuric acid freezing
kinetics are a function of film thickness and cooling rate. Thick
films freeze at lower temperatures than thin films, and the
freezing temperature decreases with increasing cooling rate. The
origin of these effects is currently under investigation, as is the
possibility that the freezing kinetics are influenced by the
underlying film substrate.
Discussion
The results presented herein establish that ultrathin films of
pure, crystalline sulfuric acid can be deposited on a Pt(111)
substrate in UHV and that the films melt incongruently at 285
K. The resulting liquids can be supercooled to ≈260 K (Tc),
at which temperature they crystallize. Also, the close cor-
respondence of the ultrathin film infrared spectra, as well as
Tm, Esub, and Evap, with those of macroscopic sulfuric acid
samples suggests that the properties of the ultrathin films are
essentially equal to those of bulk sulfuric acid. The significance
of these findings is twofold. First, they provide what is to our
knowledge the first observation of melting of a molecular solid
under ultrahigh vacuum. Detailed kinetic studies of the solid-
liquid and liquid-solid phase transitions in ultrathin sulfuric
acid films may now be carried out, as they were for the
amorphous-crystalline phase transition in ultrathin ice films.17,18
Second, by establishing a protocol for the preparation and
characterization of liquid and solid sulfuric acid in UHV, they
open the possibility of using UHV-based methods to probe
differences between the surface chemical properties of the solid
and liquid acids.
Note Added in Proof. Since this paper was submitted for
publication, a paper appeared concerning the surface composi-
tion of liquid H2SO4 + H2O solutions, as studied by Auger
electron spectroscopy and X-ray photoelectron spectroscopy.21
This work provides convincing evidence that molecular liquids
like sulfuric acid can be studied in ultrahigh vacuum.
Acknowledgment. This work was supported by the National
Science Foundation through Grant CHE-9527665.
References and Notes
(1) Hofmann, D. J.; Solomon, S. J. Geophys. Res. 1989, 94, 5029-
5041.
(2) Stuve, E. M.; Kizhakevariam, N. J. Vac. Sci. Technol. 1993, 11,
2217-2224.
(3) Dai, Q.; Hu, J.; Freedman, A.; Robinson, G. N.; Salmeron, M. J.
Phys. Chem. 1996, 100, 9-11.
(4) Schaff, J. E.; Roberts, J. T. J. Phys. Chem. 1994, 98, 6900-6902.
(5) Materer, N.; Starke, U.; Barbieri, A.; VanHove, M. A.; Somorjai,
That crystalline sulfuric acid melts incongruently to form a
mixture of H2SO4 and H2O is demonstrated by the SO3 pressure
burst at 275 K during TPD and by the release of water vapor
that accompanies conversion of the H2SO4 + H2O mixture back
to crystalline H2SO4. The driving force for incongruent melting
is minimization of the free energy. For liquid sulfuric acid at
its melting point, the minimum free energy system is a highly
concentrated H2SO4 + H2O solution (i.e., an azeotrope), the
formation of which requires the decomposition of some H2SO4
to H2O, which remains in the liquid and gaseous SO3. The
azeotropic composition in UHV was not determined in this
work, but the thermodynamic properties of highly concentrated
sulfuric acid solutions at pressures between 100 and 1000 Torr
G. A.; Kroes, G. J.; Minot, C. J. Phys. Chem. 1995, 99, 6267-6269.
(6) Brown, D. E.; George, S. M.; Huang, C.; Wong, E. K. L.; Rider,
K. B.; Smith, R. S.; Kay, B. D. J. Phys. Chem. 1996, 100, 4988-4995.
(7) Guldan, E. D.; Schindler, L. R.; Roberts, J. T. J. Phys. Chem. 1995,
99, 16059-16066.
(8) Bolsaitis, P.; Elliot, J. F. J. Chem. Eng. Data 1990, 53, 69-85.
(9) Schaff, J. E.; Roberts, J. T. J. Phys. Chem., in press.
(10) Roberts, J. T. In Proceedings of the 1995 SPIE Meeting on Laser
Techniques for Surface Science II; SPIE: Bellingham, WA, 1995; pp 125-
134.
(11) Gillespie, R. J.; Robinson, E. A. In Non-aqueous SolVent Systems;
Waddington, T. C., Ed.; Academic Press: New York, 1965; pp 117-210.
(12) Gigue´re, P. A.; Savoie, R. J. Am. Chem. Soc. 1963, 85, 287-289.
(13) Goypiron, A.; Villepin, J. d.; Novak, A. J. Chim. Phys. Phys.-Chim.
Biol. 1978, 75, 889-894.

19586 J. Phys. Chem., Vol. 100, No. 50, 1996
Schindler and Roberts
(14) Snow, K. B.; Thomas, T. F. Int. J. Mass Spectrom. Ion Phys. 1990,
96, 49-68.
(19) Kunzler, J. E. Anal. Chem. 1953, 25, 93-103.
(20) (a) Zhang, R.; Wooldridge, P. J.; Abbatt, J. P. D.; Molina, M. J. J.
Phys. Chem. 1993, 97, 7351-7358. (b) Ohtake, T. Tellus 1993, 45B, 138-
144. (c) Williams, L. R.; Long, F. S. J. Phys. Chem. 1995, 99, 3748-
3751.
(21) Fairbrother, D. H.; Johnston, H.; Somorjai, G. J. Phys. Chem. 1996,
100, 13696-13700.
(15) Yates, J. T. In Methods of Experimental Physics; Academic Press:
New York, 1985; Vol. 22; pp 425-464.
(16) Giauque, W. F.; Horning, E. V.; Kunzler, J. K.; Rubin, T. R. J.
Am. Chem. Soc. 1960, 82, 62-70.
(17) Speedy, R. J.; Debenedetti, P. G.; Smith, R. S.; Huang, C.; Kay,
B. D. J. Chem. Phys. 1996, 105, 240-244.
(18) Smith, R. S.; Huang, C.; Wong, E. K. L.; Kay, B. D. Surf. Sci.
Lett., in press.
JP962251T