The growth of silver on an ordered alumina surface

Article abstract of DOI:10.1021/jp045948k

The growth of Ag on an ordered Al₂O₃ surface was studied by low energy ion scattering spectroscopy (LEIS), scanning tunneling microscopy (STM), X-ray photoelectron spectroscopy (XPS), and temperature programmed desorption (TPD). Three-dimensional (3D) growth of Ag clusters was observed with STM and LEIS, with the cluster size increasing with Ag coverage. The XPS core level binding energies and the Auger parameters indicate a weak interaction between the Ag clusters and the Al₂O₃ support. Final state effects are determined to be the primary contribution to the Ag core level binding energy shift. Nonzero order kinetics was observed for Ag desorption in TPD with the Ag sublimation energy decreasing with decreasing cluster size. ? 2005 American Chemical Society.

Full text of DOI:10.1021/jp045948k

4064
J. Phys. Chem. B 2005, 109, 4064-4068
The Growth of Silver on an Ordered Alumina Surface
K. Luo, X. Lai, C.-W. Yi, K. A. Davis, K. K. Gath, and D. W. Goodman*
Department of Chemistry, Texas A&M UniVersity, P.O. Box 30012, College Station, Texas 77842-3012
ReceiVed: September 8, 2004; In Final Form: NoVember 24, 2004
2 3
The growth of Ag on an ordered Al O surface was studied by low energy ion scattering spectroscopy (LEIS),
scanning tunneling microscopy (STM), X-ray photoelectron spectroscopy (XPS), and temperature programmed
desorption (TPD). Three-dimensional (3D) growth of Ag clusters was observed with STM and LEIS, with
the cluster size increasing with Ag coverage. The XPS core level binding energies and the Auger parameters
indicate a weak interaction between the Ag clusters and the Al O support. Final state effects are determined
2 3
to be the primary contribution to the Ag core level binding energy shift. Nonzero order kinetics was observed
for Ag desorption in TPD with the Ag sublimation energy decreasing with decreasing cluster size.
Introduction
3 K/s. The second chamber was equipped with STM (Omicron-
-
10
1
), XPS, AES, and LEED, with a base pressure of 1 × 10
Alumina-supported silver catalysts are widely used for
Torr. The STM images were collected in a constant current
topography (CCT) mode.
1,2
3-5
ethylene oxidation, ethylene epoxidation,
formaldehyde
oxidation, methanol oxidation, and NOX reduction.9,10 The
surface morphology and the electronic/chemical properties of
Ag/alumina then are very important with respect to the function
and optimization of commercial catalysts. Model catalysts,
6
7,8
The Re(0001) single crystal was spot-welded to high-purity
tantalum wires connected to two copper leads. The sample could
be heated to 1500 K by resistive heating, to 2400 K by electron-
beam (e-beam) heating, and cooled to 90 K using liquid
nitrogen. The sample temperature was measured using a W/Re
1
1,12
prepared by growing Ag on ordered Al2O3 single crystals
or thin films on a Mo(110) substrate,13 have been studied by
temperature programmed desorption (TPD), atomic force mi-
croscopy (AFM), X-ray photoelectron spectroscopy (XPS), and
isothermal measurements. Sticking probabilities near zero at 700
(Type C) thermocouple spot-welded to the back of the crystal.
The Re crystal was cleaned by e-beam heating to 2300 K until
no contamination was detected by AES and XPS, and a sharp
1
× 1 hexagonal LEED pattern was evident.
K and relatively small sublimation energies have been reported
14
for Ag on R-Al2O3(11 2h 0).11 These authors also reported that
The alumina film was synthesized by exposing the Re(0001)
-
5
1
2
surface to Al vapor in a O2 pressure of 1 × 10
Torr at 300
Ag forms clusters on this surface as observed by AFM. In
-
6
addition, Rodriguez et al.13 reported 0.4-0.6 eV Ag 3d XPS
K with a subsequent anneal to 1000 K in O2 (1 × 10 Torr).
Afterward, the film was annealed to 1000 K in UHV, and then
characterized by XPS, AES, LEIS, LEED, and STM. XPS and
AES studies showed the film to be fully oxidized. A hexagonal
core level binding energy shifts and nonzero order desorption
of Ag from an Al2O3 film supported on Mo(110).
In the present work, scanning tunneling microscopy (STM)
and low energy ion scattering spectroscopy (LEIS) are used to
study the growth mode of Ag on ordered Al2O3 films grown
on Re(0001). From XPS, core level binding energies and Auger
parameters are used to investigate the electronic properties of
various Ag cluster sizes and interfacial reactions. The thermal
stability and bonding between Ag and Al2O3 were also studied
by XPS and TPD.
1
× 1 LEED pattern was observed for films with thicknesses
of 1.0-10 nm. STM showed the films to be exceptionally flat
with a roughness of less than a few angstroms). The film
(
thickness was estimated by attenuation of the XPS Re 4f core
level peak intensity, assuming a mean free path for the Re 4f
photoelectrons in Al2O3 of 2.8 nm.
For the Ag film deposition, a Ag doser was made of high-
purity Ag wires (99.99%) wrapped around a Ta filament. The
deposition rate of Ag, calibrated via TPD and AES of Ag on
Re(0001), was 1 monolayer (ML) per 400 s.
Experimental Section
The experiments were carried out in two ultrahigh vacuum
(
UHV) chambers. The first chamber was equipped with XPS,
Results and Discussion
Auger electron spectroscopy (AES), LEIS, low energy electron
-10
Ag/Al2O3 Growth Mode. The growth mode of metals on
oxides is of considerable importance in the synthesis of oxide-
supported metal clusters as models for studies in heterogeneous
catalysis. The three-dimensional (3D) images in Figure 1 show
the growth of 0.05, 0.5, 1.0, and 2.0 ML Ag on alumina at 300
K. At relatively small coverages, that is, 0.05 ML, Ag forms
very small clusters, ranging from 1 to 3.5 nm in diameter, with
heights of 0.4-1.2 nm, hemispherical in shape. Thereafter, the
size of the Ag clusters increases with increasing Ag coverage.
At 0.5 ML, the clusters are 4-8 nm in diameter, with heights
of 1.0-3.0 nm. At 1.0 ML, Ag clusters range from 4.5 to 9
diffraction (LEED), and TPD with a base pressure of 7 × 10
Torr. The XPS and LEIS spectra were collected using a
concentric hemispherical analyzer (PHI, SCA 10-360). For the
XPS data, the experimental error is (0.1 eV for the Ag 3d core
level and (0.2 eV for the Ag M5VV Auger transitions due to
the relatively broad peaks. The TPD experiments were carried
out using a closed-coupled (∼2 mm), line-of-sight quadruple
mass spectrometer (QMS) with a linear sample heating rate of
*
To whom correspondence should be addressed. Fax: (979) 845-6822.
E-mail: goodman@mail.chem.tamu.edu.
1
0.1021/jp045948k CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/08/2005

The Growth of Silver on an Ordered Alumina Surface
J. Phys. Chem. B, Vol. 109, No. 9, 2005 4065
Figure 1. Constant current topography (CCT) STM images (50 nm × 50 nm) of Ag on Al
2 3
O /Re(0001) at 300 K: (a) 0.05 ML; (b) 0.5 ML; (c)
1
.0 ML; and (d) 2.0 ML.
increasing Ag coverage. A feature at 906 eV appears following
deposition of Ag. In Figure 2, the area of the Ag LEIS feature,
normalized to that for 1.0 ML Ag on Re(0001), is plotted as a
function of Ag coverage. The dashed line corresponds to a two-
dimensional (or layer-by-layer) growth mechanism. The actual
data deviate from the dashed line and are more characteristic
1
5,16
of 3D Ag growth.
The LEIS and STM data are strong
evidence for a 3D-cluster growth mode for Ag on the alumina
thin film/Re(0001) at 300 K.
The growth mode of a metal on an oxide surface is determined
by minimization of the surface free energy of the oxide (γox),
the deposited metal (γm), and the interfacial free energy
1
7-19
(γm/ox).
If the γox is less than the sum of γm and γm/ox, the
oxide surface tends to maximize its exposed area, leading to a
Volmer-Weber (VW) or 3D metal growth mode. For the
converse, layer-by-layer or layer-followed-by-cluster growth
occurs. In the present studies, Ag has a surface free energy of
2
20,21
∼1.3 J/m .
The surface free energy of alumina is ap-
2
22
Figure 2. LEIS Ag peak area, normalized to that for 1.0 ML Ag on
Re(0001), as a function of Ag coverage on Al /Re(0001). The inset
proximately 0.4 J/m smaller than that of Ag. Due to the weak
interaction between Ag and alumina (described later in the text),
γm/o is estimated to be very small. Thus, to minimize the surface
free energy, the exposed alumina surface area will be maxi-
mized; that is, Ag should grow three-dimensionally on alumina.
The experimental results show this to be the case.
2 3
O
is the LEIS spectra as a function of Ag coverage.
nm, with heights of 1.2-4.2 nm. At 2.0 ML, the Ag clusters
are 4.5-10 nm in diameter, with heights of 1.7-5.0 nm. In
magnified STM images (not shown) of 2.0 ML Ag coverage,
the clusters grow randomly on the surface with no preference
for the step edges or terraces. From the STM data, a three-
dimensional (3D) growth mode of Ag clusters is clearly
observed from 0.05 to 2 ML, with the cluster density and volume
increasing with coverage.
LEIS was also used to study the Ag cluster growth mode. In
the inset of Figure 2, LEIS spectra are shown as a function of
Ag coverage. In the bottom spectrum for zero Ag coverage,
the two LEIS features at 448 and 615 eV are assigned to O and
Al. The intensity of these features decreases slowly with
Interfacial Interaction and Electronic Properties. XPS core
level binding energy shifts as a function of cluster size have
been reported for many metal/oxide systems. Differentiation of
the relative contribution of initial and final state effects to core
level binding energy (BE) shifts inevitably is an issue. In Figure
3a and b, the Ag 3d5/2 core level binding energy and Ag M5VV
Auger transition binding energy are plotted as a function of Ag
coverage at 300 K. Both Ag 3d5/2 and M5VV Auger binding
energies shift to a higher BE with decreasing Ag coverage. At
0.04 ML Ag coverage, the Ag 3d core level and M5VV Auger
peak positions are higher than the bulk Ag value by 0.9 and

4066 J. Phys. Chem. B, Vol. 109, No. 9, 2005
Luo et al.
Experimentally, due to the wide application of Al and Mg
KR X-ray sources, most Auger transition peaks come from
C′VV or C′C′′V excitation, rather than C′C′′C′′′ excitation.
Validation of the two assumptions in eq 2 for transitions
involving the valence band strongly determines the limitation
3
5
37
of its application. Moretti and Wertheim pointed out that
the validation of the two assumptions is fully dependent on the
structure and depth of the valence band. In the case of atomic-
like Auger transitions, like O KVV, Pt MVV, or Sn MNN, etc.,
the assumptions are valid. Nevertheless, due to the large
delocalization of valence band and changes in screening effects
between core holes and valence holes, some band-like Auger
transitions should not be used to extract the Auger parameter.
In our study of the growth of Ag on ultrathin Al2O3 films (∼30
Å) using a Mg KR source, Ag has an intense and sharp Auger
M5VV (3d-4d) transition feature in the XPS spectra. As a 4d
metal, Ag has been studied experimentally and theoretically in
terms of its validity with respect to the Auger parameter
1
6,28,38,39
28
approximation.
In Ag growth on amorphous carbon
1
6
and TiO2(110) single-crystal surfaces, the core level binding
energies shift systematically with the centroid of the 4d valence
band. In the present work, the valence band was monitored (not
shown) and its centroid was observed to shift in concert with
the core level features. From Kleiman et al.’s work,3 the 4d
valence holes of all 4d metals (except Pd) have the same
relaxation in the final state as the 3d core holes. Therefore, both
assumptions in eq 2 are valid for Ag M5VV.
8,39
5
Figure 3. (a) XPS Ag 3d5/2 core level binding energy; (b) Ag M VV
Auger peak binding energy; (c) Auger parameter; and (d) initial state
contribution to the core level binding energy shifts as a function of Ag
Using eq 1, the Auger parameter (R) can be calculated from
the data in Figure 3a and b and is plotted as a function of Ag
coverage in Figure 3c. The Auger parameter decreases with
decreasing Ag coverage and shifts in the opposite direction from
the 3d core level and MVV transitions. The shift is on the order
of ∼1.6 eV. According to eq 2, the final state effect contribution
to the core level BE shift, that is, half of the change of the
Auger parameter (∆R/2), can be estimated from the data in
Figure 3c. Subtracting the final state contribution from the core
level BE shift, the initial state effect contribution can be
calculated and is plotted as a function of Ag coverage in Figure
2 3
coverage on Al O /Re(0001) at 300 K.
2.5 eV, respectively. At 10.0 ML Ag coverage, each approaches
the bulk Ag value.
Regarding the initial and finial state contributions to the core
level BE shifts, Mason, Bagus, and Parmegiani concluded that
initial state effects, such as the atomic coordination or metal-
surface interaction, dominate the shifts for small metal clusters
on poorly conducting surfaces.2
3-25
However, Egelhoff and
Tibbetts suggested that final state screening, that is, relaxation
effects, is the main cause of binding energy shifts for metal
3
d. Essentially the initial state contribution to the Ag core level
clusters with varying size.2
6,27
BE shift (0.9 eV) is very small (within (0.1 eV) from very
low (0.04 ML) to high (10.0 ML) Ag coverages. These data
clearly delineate the importance of final state contributions to
core level BE shifts. In addition, in the study of the alumina Al
Wertheim claimed that Coulombic
effects in the final state contribute primarily to the binding
energy shift for metal clusters on poorly conductive substrates
such as amorphous carbon.2
8-30
2
s and O 1s (not shown), the binding energies show no shifts
In the early 1970s, Wagner first introduced the concept of
the Auger parameter (R)31,32 to study photoemission binding
until 2.0 ML Ag. Up to 10.0 ML, both Al 2s and O 1s shift
toward a lower binding energy by ∼0.2 eV, which is interpreted
energy shifts. The Auger parameter that is widely used currently
4
0
3
3-35
as metal cluster screening on the alumina film. These results
indicate a very weak interaction between Ag clusters and the
Al2O3 thin film and suggest that the origin of the Ag core level
BE shift is a final state effect.
is defined
as
R ) E (C′) + E (C′C′′C′′′)
(1)
b
k
Another important approach for differentiating initial and final
state effects was shown in the work of Moretti.35 Using an
electrostatic model, the assumptions of similar initial state
effects, and the approximations used to derive eq 2, a plot of
Auger kinetic energy Ek(C′C′′C′′′) versus core level binding
where Eb(C′) is the core level (C′) binding energy and
Ek(C′C′′C′′′) is the kinetic energy of the Auger transitions from
the C′, C′′, and C′′′ core levels, respectively. By assuming equal
binding energy shifts and identical relaxation mechanisms for
all three core levels, the change in the Auger parameter (∆R)
is approximately twice the final state effects (∆R),3
6,37
energy E (C′) obtained from metal XPS spectra for different
including
b
size of clusters, that is, a Wagner plot, should have a linear
extra-atomic relaxation and the Coulombic energy, and is written
as
3
5,36
relationship with a slope of -3.
This method can be used
to differentiate the initial and final state effects in core level
BE shifts. Because the assumptions for eq 2 are all met for Ag,
the Wagner plot is also employed here to study the initial
and/or final state effect contributions to the binding energy
shifts. In Figure 4, the Ag M5VV kinetic energy is plotted as a
function of Ag 3d5/2 binding energy. The linear fit (solid line)
∆
R ) 2∆R
(2)
This equation offers a direct way to evaluate the contribution
of final state effects, and thus to separate initial and final state
contributions to the observed binding energy shifts.

The Growth of Silver on an Ordered Alumina Surface
J. Phys. Chem. B, Vol. 109, No. 9, 2005 4067
Figure 4. A Wagner plot of Ag M
5
VV kinetic energy versus Ag 3d5/2
Figure 6. Cluster sublimation energies (Esub) deduced using the leading
binding energy as a function of Ag coverage.
edge analysis of TPD spectra of Figure 5.
Figure 5. TPD spectra of Ag as a function of Ag coverage.
Figure 7. XPS Ag 3d5/2 peak intensity as a function of temperature
2 3
for a 0.1 ML Ag/Al O /Re(0001). The inset is the corresponding TPD
spectrum at the same Ag coverage.
yields a straight line with a slope of -2.9, which is very close
to the dashed line with a slope of -3. The similarities in
the slopes of the solid and dashed lines strongly support the
origin of the core level BE shifts in Figure 3a being final state
effects.
Thermal Sintering. Thermal sintering of oxide-supported
metal clusters has been reported for various metal-oxide
1
6,19,40,42
systems.
In the present study, this effect was studied
Temperature Programmed Desorption of Ag. Figure 5
shows a series of TPD spectra of Ag on Al2O3/Re(0001) as a
function of the Ag coverage (0.1-1.8 ML). The peak profiles
are asymmetric with the leading edges moving toward higher
temperatures with increasing Ag coverage (shown in the inset).
On the low-temperature side of the peaks, the desorption rate
is higher for low coverages as compared to high coverages. On
the high-temperature side, the desorption rate drops sharply. Our
with XPS and TPD. 0.1 ML Ag was deposited on the Al2O3
film at 100 K, annealed, after which XPS spectra were collected
at 100 K. In Figure 7, the XPS Ag 3d5/2 peak area intensities
are plotted as a function of temperature. The inset is the TPD
spectrum of the corresponding Ag coverage. From 100 to 500
K, the Ag peak intensity decreases slightly, consistent with
cluster sintering. From 500 to 800 K, although the temperature
is still well below the desorption temperature of Ag, the peak
intensity falls rapidly showing that significant sintering of the
Ag clusters occurs well before desorption. After 800 K, the peak
intensity falls concurrently with the desorption of Ag until the
surface is Ag-free.
results are consistent with studies of Ag on R-Al2O3 single
crystals and alumina thin films on Ru and Mo1
1,13
in that the
order of the desorption is fractional rather than zero. In Monte
41
Carlo simulations of the TPD spectra, the various metal cluster
sizes are considered as finite islands. With increasing coverage,
the average cluster size increases and the peak maximum shifts
toward higher temperature. At a certain temperature, the
desorption rate increases significantly and desorption occurs
within a narrow temperature range. Thereafter, the desorption
rate falls rapidly after reaching the peak desorption temperature,
giving rise to a sharp drop in the desorption rate on the high-
temperature side of the desorption peak.
To investigate the interfacial interactions upon annealing, the
Ag 3d5/2 binding energy, Ag M VV binding energy, Auger
5
parameter, and initial state contribution to Ag core level BE
shift at 0.1 ML Ag were studied. The same methodology used
in Figure 3 was employed with the anneal temperature as the
variable rather than Ag coverage. It was observed that Ag core
level and Auger transition binding energies shift to lower binding
energies with increasing temperature, reaching the bulk value
near 800 K. The Auger parameter, on the other hand, shifts
toward higher binding energies with increasing temperature. The
initial state contribution to the core level BE shift, caused by
the temperature-induced sintering, is less than ∼0.1 eV. Ad-
ditionally, a Wagner plot (not shown) of the Ag M5VV kinetic
energy versus the Ag 3d5/2 binding energy indicates a linear
relationship with a slope of -3.1. No large binding energy shifts
were observed on either O or Al core level peaks. Therefore,
all of the data indicate that the temperature-induced core level
BE shifts are due primarily to final state effects.
The sublimation energies (Esub) of the Ag clusters as a
function of coverage, extracted using the leading edge analysis,
are plotted in Figure 6. The Esub’s drop to approximately 20
kcal/mol at a Ag coverage of 0.2 ML. The sublimation energy
increases with increasing coverage until the bulk value is reached
at approximately 3.0 ML. The increase in the number of low-
coordinated Ag atoms in small clusters relative to larger clusters
combined with a stronger Ag-Ag as compared to Ag-Al2O3
interaction likely are responsible for the large variation in the
Ag sublimation energies.

4068 J. Phys. Chem. B, Vol. 109, No. 9, 2005
Luo et al.
Conclusions
(15) Diebold, U.; Pan, J.-M.; Madey, T. E. Phys. ReV. B 1993, 47, 3868.
(
16) Luo, K.; St. Clair, T. P.; Lai, X.; Goodman, D. W. J. Phys. Chem.
As an important model catalyst, Ag/Al2O3 ultrathin film/
Re(0001) was systematically studied by STM, LEIS, XPS, and
TPD. At 300 K, three-dimensional Ag growth was observed.
The use of the Auger parameter and Wagner plots allows
separation of initial and final state effects in core level binding
energy shifts for Ag clusters of various sizes. Final state effects
are concluded to be primarily responsible for the core level
binding energy shifts. No strong interfacial interaction between
Ag and alumina was apparent between 100 K and the desorption
temperature of Ag. TPD data indicate weak bonding between
Ag and alumina within the small cluster regime.
B 2000, 104, 3050.
(17) Bauer, E. Z. Kristallogr. 1958, 110, 372.
(18) Egelhoff, W. F., Jr. In Ultathin Magnetic Structures I, An
Introduction to Electronic, Magnetic, and Structural Properties; Bland, J.
A. C., Heinrich, B., Eds.; Springer-Verlag: Berlin, 1994.
(19) Persaud, R.; Madey, T. E. In Growth and Properties of Ultrathin
Epitaxial Layers; King, D. A., Woodruff, D. P., Eds.; Elsevier Science B.
V.: New York, 1997.
(
(
20) Mezey, L. Z.; Giber, J. Jpn. J. Appl. Phys. 1982, 21, 1569.
21) Vitos, L.; Ruban, A. V.; Skriver, H. L. Philos. Mag. B 1998, 78,
4
87.
(22) Overbury, S. H.; Bertrand, P. A.; Somorjai, G. A. Chem. ReV. 1975,
5, 547.
7
(
(
23) Mason, M. G. Phys. ReV. B 1983, 27, 748.
24) Parmigiani, F.; Kay, E.; Bagus, P. S. J. Electron Spectrosc. Relat.
Acknowledgment. We would like to acknowledge support
of this work by the Department of Energy, Office of Basic
Energy Sciences, Division of Chemical Sciences, and the Robert
A. Welch Foundation.
Phenom. 1985, 36, 257.
25) Bagus, P. S.; Brundle, C. R.; Pacchioni, G.; Parmegiani, F. Surf.
Sci. Rep. 1993, 19, 265.
(
(
(
26) Egelhoff, W. F., Jr.; Tibbetts, G. G. Phys. ReV. B 1979, 19, 5028.
27) Egelhoff, W. F., Jr.; Tibbetts, G. G. Solid State Commun. 1979,
2
9, 53.
28) Wertheim, G. K.; DiCenzo, S. B.; Youngquist, S. E. Phys. ReV.
Lett. 1983, 51, 2310.
References and Notes
(
(
1) Corner, P. T.; Kovenklioglu, S.; Shelly, D. C. Appl. Catal. 1991,
(
29) Wertheim, G. K.; DiCenzo, S. B.; Buchanan, D. N. E. Phys. ReV.
7
1, 247.
B 1986, 33, 5384.
(
(
(
2) Mao, C.-F.; Vannice, M. A. Appl. Catal., A 1995, 122, 61.
3) Minahan, D. M.; Hoflund, G. B. J. Catal. 1996, 158, 109.
4) Hoflund, G. B.; Minahan, D. M. Nucl. Instrum. Methods Phys. Res.,
(30) Wertheim, G. K. Z. Phys. D: At., Mol. Clusters 1989, 12, 319.
(31) Wagner, C. D. Anal. Chem. 1972, 44, 967.
Sect. B 1996, 118, 517.
5) Bukhtiyarov, V.; Prosvirin, I. P.; Kvon, R. I.; Goncharova, S. N.;
Bal’zhinimaev, B. S. J. Chem. Soc., Faraday Trans. 1997, 93, 2323.
(32) Wagner, C. D. Faraday Discuss. Chem. Soc. 1975, 60, 291.
(33) Gaarenstroom, S. W.; Winograd, N. J. Chem. Phys. 1977, 67, 3500.
(
(34) Wagner, C. D.; Gale, L. H.; Raymond, R. H. Anal. Chem. 1979,
(
6) Mao, C.-F.; Vannice, M. A. J. Catal. 1995, 154, 230.
5
1
1, 466.
(7) Wang, J.-A.; Aguilar-Rios, G.; Wang, R. Appl. Surf. Sci. 1999,
(
(
35) Moretti, G. J. Electron Spectrosc. Relat. Phenom. 1998, 95, 95.
36) Wu, Y.; Garfunkel, E.; Madey, T. E. J. Vac. Sci. Technol., A 1996,
1
47, 44.
(8) Cordi, E. M.; Falconer, J. L. Appl. Catal., A 1997, 151, 179.
9) Takagi, K.; Kobayashi, T.; Ohkita, H.; Mizushima, T.; Kakuta, N.;
4, 1662.
(
(
(
37) Wertheim, G. K. Phys. ReV. B 1987, 36, 9559.
38) Kleiman, G. G.; Landers, R.; Nascente, P. A. P.; de Castro, S. G.
Abe, A.; Yoshida, K. Catal. Today 1998, 45, 123.
10) Abe, A.; Aoyama, N.; Sumiya, S.; Kakuta, N.; Yoshida, K. React.
Kinet. Catal. Lett. 1998, 65, 139.
(
C. Phys. ReV. B 1992, 46, 4405.
39) Kleiman, G. G.; Landers, R.; Nascente, P. A. P.; de Castro, S. G.
C. Surf. Sci. 1993, 287/288, 798.
40) Luo, K.; Kim, D. Y.; Goodman, D. W. J. Mol. Catal. A 2001, 167,
91.
(41) Masel, R. I. Principles of Adsorption and Reaction on Solid
(
(
11) Van Campen, D. G.; Hrbek, J. J. Phys. Chem. 1995, 99, 16389.
(12) Beitel, G.; Markert, K.; Wiechers, J.; Hrbek, J.; Behm, R. J. In
(
Spring Series in Surface Science; Umbach, E., Freund, H.-J., Eds.; Springer-
Verlag: Berlin, Heidelberg, 1993; Vol. 33.
1
(
13) Rodriguez, J. A.; Kuhn, M.; Hrbek, J. J. Phys. Chem. 1996, 100,
8240.
14) Lai, X.; Chusuei, C. C.; Luo, K.; Guo, Q.; Goodman, D. W. Chem.
Phys. Lett. 2000, 330, 226.
1
Surfaces; Wiley Series in Chemical Engineering; Wiley: New York, 1996.
(42) Zhang, L.; Persaud, R.; Madey, T. E. Phys. ReV. B 1997, 56,
10549.
(