AVS2001 Session SS2-MoP: Surfaces and Interfaces Poster Session

Monday, October 29, 2001 5:30 PM in Room 134/135
Monday Afternoon

Time Period MoP Sessions | Topic SS Sessions | Time Periods | Topics | AVS2001 Schedule

SS2-MoP-1 Effects of Na and Se Annealing on CuInSe2 Single Crystals
V. Lyakhovitskaya, K. Gartsman, Y. Feldman, H. Cohen, D. Cahen (Weizmann Institute of Science, Israel)
We report on X-ray photoelectron spectroscopic, secondary ion mass spectroscopy, scanning electron microscopy, electron-beam induced current and energy dispersive X-ray fluorescence studies of high quality, travelling heater-grown single crystals of CuInse2, that were subjected to controlled heat treatments from 300 C upwards for varying times, in Se atmosphere, in Na atmosphere and in a mixture thereof. We find that with Na the surface of the crystals is gradually destroyed, while with Se a gradual topotactic type conversion takes place, controlled by Cu diffusion towards the surface, rather than by Se diffusion into the bulk. This is accompanied by formation of a separate Cu-Se phase on the surface. The implications of these findings for the effects of Na on the performance of CuInse2-based solar cells will be considered, in terms of both possible surface and bulk effects.
SS2-MoP-2 Two-dimensional and Three-dimensional Growth of Ag Islands on a Ge(001) Surface Studied by Photoemission Spectroscopy
M. Yamada, K. Nakatsuji, T. Iimori, S. Ohno (ISSP, The University of Tokyo, Japan); Y. Naitoh (CREST, Japan Science and Technology Corp.); T. Okuda, A. Harasawa, T. Kinoshita, F. Komori (ISSP, The University of Tokyo, Japan)
Growth mode of Ag thin films on a Ge(001) surface depends on the substrate temperature: two-dimensional (2D) growth is predominant at low temperatures and three-dimensional (3D) growth becomes dominant at room temperature.1 According to the recipe found by the STM study, we prepared samples composed of 2D (monolayer-thick) or 3D (several-monolayer-thick) Ag islands, and investigated their electronic structure by synchrotron radiation photoemission spectroscopy. A clean Ge(001) surface gives a Ge3d spectrum with a shoulder at the low binding energy side, which is due to surface core-level shift (SCLS) of topmost, 2x1 reconstructed Ge atoms. After Ag deposition at 95K, the shoulder disappeared and the LEED pattern changed to 1x1. These facts indicate that the interaction between Ag and Ge atoms is strong enough to destroy the 2x1 reconstruction in the 2D Ag islands. The Ag4d peak was narrow until Ag thickness of 1.3Å, indicating that the Ag-Ag interaction is weak in the 2D Ag islands. Above 1.3Å a new broad component was observed in the Ag4d spectrum, which suggested the presence of 3D Ag islands. The breadth of the new component indicates the strong Ag-Ag interaction in the 3D Ag islands. When Ag was deposited on a Ge(001) surface at room temperature, the LEED pattern remained to be 2x1 and the shoulder due to the SCLS did not disappear in the Ge3d spectrum. The Ag4d band was broad and bulk-like. These facts indicate that Ag grows three-dimensionally and leaves intact most of the clean Ge(001)-(2x1) surface at room temperature.

1 F. Komori et al., Surf. Sci., 438 (1999) 123.

SS2-MoP-3 Initial Growth of Ag Islands on Ge(001) Surface at Low Temperature
Y. Naitoh, K. Nakatsuji, F. Komori (ISSP, University of Tokyo, Japan); K. Seino, A. Ishii (Tottori University, Japan)
The Ag on Ge(001) surface has been an interesting issue for the last decade since suddenly increase of the surface electronic conductivity at 2K was reported.1 It is suggested the interface between Ag and Ge has peculiar electronic state because neither Ag nor Ge bulk have superconductivity. In the latest STM observation2 showed that mono-atomic-layer (ML) thick Ag islands parallel to the dimer-row of the Ge(001) surface and Ag chains along the dimmer direction are formed at the initial stage of the Ag deposition on a 90 K substrate followed by annealing to room temperature (RT). In the present paper, we show the Ag adsorbed sites on Ge(001) surface by STM without annealing to RT. The STM experiments were performed in an ultra high vacuum (below 1x10-8Pa) system. The surface of a Ge(001) crystal, which was cut from n-type Ge(001), was cleaned by several cycles of Ar+ bombardment and annealing. Ag was deposited onto the clean Ge surface at 90K with the rate of 1 ML/min, and then the substrate was moved to the STM unit as keeping the low temperature. Bright dots, which indicate Ag atoms, were observed between the Ge dimer rows. It was found some dots were arranged to form chains along both the dimer row and the dimer direction having intervals of 2aGe (aGe=0.4nm) and islands with 2aGe width elongating along the dimer row. As annealing the substrate to RT bright dots disappeared and the conventional Ag chains and islands were found to be formed. This shows Ag sites on Ge(001) surface without annealing to RT are quasi-stable. These were compared with the stable site of Ag on the surface given by the first-principles calculation.

1 M. J. Burns et al., Solid State Comm. 51, 865 (1984)
2 F. Komori et al., Surf. Sci. 438, 1234 (1999).

SS2-MoP-4 In-Situ X-Ray Reflectometry of the Martensitic Transformation of AuCd Alloys
H. Shimazu, S. Shibata, S. Doi, I. Takahashi (Kwansei Gakuin University, Japan); T. Ohba (Shimane University, Japan)
Numerous works have been dedicated to the study of meaning of surfaces for bulk phase transitions. For second order transitions, importance of the dimensionality has been understood thoroughly. Thus, an answer for this issue might have been obtained. Compared to such a situation, we must admit that many phenomena at the surfaces for first order phase transitions have not fully been understood. Especially, peculiarity of nucleation and evolution of domains at the surface triggered by the first order bulk transition remain unveiled, although those in bulk have been investigated. The martensitic transformation is a first order phase transition, responsible for shape-memory effects. In a low temperature phase, bulk domains called "variants" appear. On the other hand, the surface exhibits a morphological change (surface relief). Many people consider that the surface relief is a mere termination of the bulk variants. If they are correct, the evolution of the surface relief may start simultaneously as the variants start to evolve. However, as far as the authors' knowledge, there is no accepted study on the relationship between nucleation of the surface relief and that of the bulk variants. In the present paper, we focus on in-situ observations about the surface relief and the micro-variants in the vicinity of martensitic transformations of AuCd alloys: We observed the surface relief by X-ray reflectivity, whereas Bragg reflection gave information on bulk variants.1 An image plate detector was adopted with an X-ray camera.2 Our study revealed peculiar properties on the evolution of the surface relief. They indicated that the surface plays an intrinsic role at the first stage of the martensitic transformation. An aging effect was also detected in the surface diffraction.

1 S. Doi and I. Takahashi; Philosophical Magazine A, 80(2000) 1889-1899.
2 S. Shibata, S. Doi and I. Takahashi; to be published in Surface Science (2001).

SS2-MoP-6 Metastable Crossed Row Structure and Nano-structural Change on TiO2(001) Studied by STM
R. Tero, K. Fukui, Y. Iwasawa (The University of Tokyo, Japan)
Atom-resolved structures on a rutile TiO2(001) surface were visualized by scanning tunneling microscopy (STM). We have found that a flat ordered structure, which consists of crossed rows along the [110] and the [11-0] directions, can be prepared by annealing an Ar ion-sputtered TiO2(001) surface at 850-1050 K under UHV with heating and cooling rates at 7-10 K s-1. Its LEED pattern was identical to that for a {114}-faceted structure proposed in previous studies. The rows had a bleacher-like structure, which formed stairs at both sides with an average slope identical to the {114} face. An atom-resolved STM image showed that units of three bright spots, which were lined perpendicular to the steps, were arranged by 0.65 nm on each narrow step. The crossed row structure was metastable and its stability depended on the degree of the bulk reduction. Structural transformation occurred at much lower temperatures than that for preparing the crossed row structure. Particles with nm scale appeared on the crossed row structure by reannealing the surface under UHV at 400-490 K. The surface was covered up with the particles of 0.9-6 nm in diameter. Adsorption of formic acid on the crossed row structure before reannealing drastically modified the structural transformation. The particle size was distributed to 2-8 nm in diameter after the formic acid-covered crossed row structure was reannealed at 710 K. The stability of the crossed row structures and the mechanism of the particle growth, which depended on the degree of bulk reduction and reaction of adsorbates, will be discussed.
SS2-MoP-7 Theoretical Analysis of Short Range Order of Au/Si(100) 5x3.2 Structure
Y. Yagi, K. Kakitani, H. Kaji, A. Yoshimori (Okayama University of Science, Japan)
It is known that the metal adsorbed semiconductor surfaces show various characteristic structures. One of them is the Au adsorbed on Si(100) surface. According to the recent experimental results by STM and LEED, this system shows the 5x3.2 structure.1-4 Au adsorbates and Si substrate atoms form the 5x3 structure. This structure is made up with the "band" structure along the direction of 3 times and the "band" structures arrange with 5 times period in the perpendicular direction. The STM results indicate that the 5x3.2 structure seems to be formed by protrusions called "wall" in the topmost layer on the 5x3 structure. We describe this system by the lattice gas model for adsorbates (adatoms) in the topmost layer on the 5x3 structure with semi-long ranged repulsive interaction and short ranged attractive interactions, assuming that the protrusion is formed by the adsorbate (adatom), and perform Monte Carlo simulation. We obtain the two adsorbate positional correlation functions and the structure factor of this surface system. The obtained theoretical results are compared with experimental ones, in particular, the correlation function obtained from the STM images. Our results reproduce experimental features semi-quantitatively. The degree of ordering in the 3.2 times period does not seem perfect in the STM images and it may be a structure with short range order. The ratio of interactions with temperature and the most probable interaction ratios are determined by the comparison.

1X. F. Lin, K. J. Wan, J. C. Glueckstein and J. Nogami, Phys. Rev. B47, (1993) 3671.
2G. Jayaram and L. D. Marks, Surf. Rev and Lett., 2 (1995) 731.
3H. Minoda, K. Yagi, F.-J. Meyer zu Heringdorf, A. Meier, D. Kahler, and M. Horn von Hoegen, Phys. Rev. B59, (1999) 2363.
4R. Hild, F.-J. Meyer zu Heringdorf, P. Zahl and M. Horn-von Hoegen, Surf. Sci. 454-456 ( 2000) 851. .

SS2-MoP-9 Experimental and Theoretical Investigation of Multi-layer Films with Ultra-thin Intermediate Layers
I.G. Levchenko (Kharkov Aerospace University, Ukraine)
The multi-layer films were applied by the ion deposition method. The three-layered and four-layered films consisting of titanium carbide, titanium nitride, chromium nitride, zirconium nitride, carbonitrides, and aluminum oxide with ultra-thin copper and silver intermediate layers were investigated. The total film thickness was ten to twenty five micrometers. The ultra-thin layers were applied between the main layers, the ultra-thin layer thickness being of ten to hundred atomic layers. The main layers were applied at a voltage of one or two hundred volts and the intermediate layers were applied at five to eight hundred volts. The conditions of intermediate layer application ensured creation of the continuous layers. The aim of ultra-thin layers was to provide relaxation in the main layers, to remove the residual stresses, to ensure the high level of adhesion force between the main layers, and to reduce the porosity due to the penetration of ultra-thin layer metal into the main layers. The measurements of the porosity coefficient were performed by electrolytic method using the hydrogen penetration. The porosity factor was found to decrease twice for copper intermediate layers and thrice for the silver intermediate layers. The micro-stress levels were found to be five times less than those for the film of the equal thickness applied without the intermediate layers. The abrasion test had shown the improved wear resistance due to the absence of microchipping as a result of low micro-stress level and high enough film thickness. The main layer application voltage - intermediate layer application voltage ratio was found to be the significant value in reducing a porosity factor.
SS2-MoP-10 About the Interface between a Quasicrystal and its Oxide Overlayer
V. Fournèe, J.W. Anderegg, A.R. Ross, T.A. Lograsso, P.A. Thiel (Iowa State University)
When the surface of an Al-Pd-Mn icosahedral quasicrystal (QC) is exposed to air at room temperature, an almost pure alumina overlayer develops across it. The thickness of this oxide is rather small and is similar to the passivating layer on fcc-Al. This is of course an attractive characteristic which, combined with other surface properties like low friction coefficients or low adhesion, makes QC surfaces relevant for technological applications.1 In particular, the low adhesion property, related to a low surface energy, is believed to be influenced by the gradient, through the surface, of the electronic density of states at the Fermi level (N(EF)). This assume that the low N(EF) characteristic of the bulk electronic structure of QC is effectively maintained up to the surface. However, in a practical situation where an aluminium oxide overlayer is formed, the interface just under the oxide may be depleted in Al and enriched in Pd and Mn. The concentration may be shifted out of the stability range of the QC phase and structural transformation may occur. This also should affect the electronic properties at the interface. In an effort to clarify this point, we will present our recent experimental results on this subject. Experiments were performed on a five-fold surface of a single grain Al-Pd-Mn icosahedral QC. The clean surface was prepared in UHV and later oxidized in-situ. A depth profile of the chemical composition across the oxidized surface is acquired both by Auger and photoelectron spectroscopies. Informations on the morphology of this surface are derived from both scanning electron and scanning tunneling microscopies. Finally, an analysis of the photoelectron core-level lineshape, as a function of the escape depth of the photoelectrons, provides insight into the variations of the metallic character through the surface.

1 P.A. Thiel, C. J. Jenks, A. I. Goldman, in Physical properties of Quasicrystals, ed. Z. Stadnik (Springer Verlag, Berlin, 1999).

SS2-MoP-11 Temperature Effect of Well-ordered Al2O3 Film on NiAl(110)
T.T. Lay, M. Yoshitake, B. Mebarki (National Institute for Material Science, Japan)
It is known that the oxidation of the (110) surface of intermetallic compound NiAl leads to the formation of a well-ordered Al2O3 film.1 It has the potential of application in metal-insulator-metal(M-I-M) electron emitter which needs wide gap insulator. In our previous work epitaxial ultra-thin Al2O3 film was grown on NiAl(110) by introducing 1200L oxygen at 570K and subsequent annealing at 1070K. The crystalinity of the epitaxial film varied from pressure 5 x 10-7 to 5 x 10-8 Torr. The best crystalinity was obtained when oxygen was introduced at pressure of 5 x 10-8 Torr. However, to get 1200L oxygen at 5 x 10-8 Torr, the time needed is 400minutes. For thick epitaxial Al2O3 film, multi-oxidation is required and time needed to growth desire film thickness is too long. To get well-ordered thick expitaxial Al2O3 film in higher growth rate, we have carried out the temperature control in oxidation. Oxidation temperature varied from 570K to 720K at constant pressure of 5 x 10-7 Torr for 1200L. After oxidation subsequent annealing is performed at 1070K in a vacuum chamber equipped with LEED/AES. From the LEED pattern, the crystalinity of epitaxial Al2O3 film become distinct as temperature increases and shows optimum at temperature around 670K. AES measurement also has good correlation with the results of LEED patterns. The ratio of O/Ni peak intensity is directly related to film thickness.2 The optimum ratio of O/Ni was obtained at temperature around 670K and approximately equal to several multi-oxidations at pressure 5 x 10-8 Torr at 570K. The result shows crystalinity and thickness can be control by oxidation temperature.

1 R.M. Jaeger, H.Kuhlenbech, H.J.Freund, M.Wuttig, W. Hoffmann, R. Franchy and H. Ibach, Surf. Sci., 259,235(1991)
2 M. Yoshitake and B. Mebarki, 6th. Int.Symposium on Advanced Physical Fields, Tsukuba, Japan, March 6-9, 183(2001)

SS2-MoP-12 Initial Growth and Alloy Formation of Ni Ultrathin Films on Pt(111)
C.W. Su, Y.E. Wu, C.S. Shern (National Taiwan Normal University)
Low-energy electron diffraction (LEED), Auger electron spectroscopy (AES) and Ultraviolet photoelectron spectroscopy (UPS) were used to study of the initial growth of Ni ultrathin films on Pt(111) at room temperature. As the coverage of Ni between 0.3 ML to 1.0 ML , a (√(3)*√(3))R30° commensurate structure was observed. As the coverage increases from 1 ML to 1.2 ML, the superposition of Pt(1x1) and epitaxial Ni(1x1) LEED pattern is formed. When the coverage of Ni increases from 1.3 ML to 6 ML, the structure is changed to the incommensurate epitaxy with fine satellites around the (1x1) hexagonal LEED pattern. Deposition of Ni films more than 7 ML up to 12 ML, the p(2x2) LEED pattern is appeared gradually. After high temperature annealing, the formation of Ni-Pt alloy with different coverages of Ni was confirmed by AES and UPS. The starting temperature of alloy formation increases as the coverage of Ni increases. The density of state also shows some interesting changes during the thermal annealing process.
SS2-MoP-14 Photoemission from Graphite Overlayers on SiC(0001)
T. Balasubramanian (Lund University, Sweden); T. Kihlgren, L. Walldén (Chalmers University of Technology, Sweden); R. Yakimova (Linköping University, Sweden)
For lack of large graphite crystals with well defined termination overlayers of the material prepared by heating SiC can provide an attractive alternative to natural crystals or to samples of highly oriented pyrographite, which is often used as substrate for adsorption and absorption studies. Using LEED, STM and UPS (20 to 140 eV photon energy) for sample characterization we find that heating (2200 K, 1/2 h) an off axes (8 degrees) 4H-SiC(0001) produces a well ordered, multilayer thick overlayer, which shows a 3-fold LEED pattern and gives quite well resolved photoemission spectra. We use the latter to monitor the dispersion of valence electrons, photoemission line widths and cross sections. Particularly narrow emission lines are observed for the Fermi level electrons near K-H in the Brillouin zone and for the upper σ state at the zone center. Of interest for adsorption studies is that the latter gives a line width on par with that of the C 1s core level.For an on axes 6H-SiC(0001) crystal a similar preparation procedure yields a 6√3x6√3 LEED pattern and photoemission spectra with no dispersion along the c-axes for the π state, which indicates that the overlayer in this case is a single graphene layer. Such a layer in vacuum should be a zero gap semiconductor with a band crossing at K defining the Fermi level. For the adsorbed monolayer we find that the bands do not cross but form a 0.2 eV gap centered 0.35 eV below the Fermi energy. The gap splits of a thin electron gas with a filled band width of 0.25 eV and an estimated surface density corresponding to 0.06 electrons per C atom.For both mono- and multilayers the photoemission cross sections are oscillatory with nearly out of phase variations for the π and σ states when plotted versus the kinetic electron energy and the curves shifted to around 5 eV higher energy for the monolayer.
SS2-MoP-15 Local Tunneling Barrier Height Measurement of Cs Monolayer on a Pt(111) Surface
Y. Yamada, A. Sinsarp, M. Sasaki, S. Yamamoto (University of Tsukuba, Japan)
The local tunneling barrier height (LBH) image of the Cs monolayer on a Pt(111) surface is measured in atomic resolution by means of an STM technique. The Cs atoms form three different superstructures, (2x2), (√3x√3) and (√3x√3)R30, depending on its surface coverage on the Pt(111) surface. We find that the (2x2) structure is thermally stable at room temperature in contrast to other structures which disappear in few hours. Only for the (2x2) structure, we could obtain a clear atom-resolved STM/LBH image at room temperature. We find the atomic LBH distribution in the image corresponding to atomic geometry of this structure. This distribution, however, cannot be considered to be identical to the distribution of the classically defined work function. The average barrier height of this structure is, however, about 2.2eV smaller than that of a clean Pt(111) surface, suggesting the work function reduction of about 4.4eV from the clean Pt(111) value. This agrees well with our macroscopical work function results measured by means of the Kelvin contact potential method. The (2x2) surface corrugation amplitude obtained from the STM image is as large as 0.6Å, suggesting that the spatial distribution in the electron density of this structure is strongly localized, in contrast to the case of other 'metallic' superstructures, where no atomic image is observed by STM. It is speculated that the charge localization with some charge transfer causes the lowest work function of the (2x2) structure and its high thermal stability.
SS2-MoP-16 Dynamics and Thermal Stability of Cs Super-structures on a Pt(111) Surface
T. Kondo, H. Kozakai, T. Sasaki, S. Yamamoto (University of Tsukuba, Japan)
Understanding the dynamics and the thermal stability of alkali-metal atoms adsorbed on a metal surface is essentially important for controlling the monolayer growth and unveiling the origin of the work function reduction due to adsorption. We have made clear the dynamics from formation to destruction and the thermal stability of the Cs super-structures on a Pt(111) surface by means of Helium atom scattering (HAS) which enables us to study at high sensitivity the dynamics of the ultra-thin delicate layer of Cs atoms. Among three structures previously identified by LEED, both the (√3x√3) and the (√3x√3)R30° structures are found unstable and destructed at around 350K, transforming into the more stable (2x2)structure. The (2x2) structure which is stable up to about 450K exhibits a large corrugation of the potential energy surface seen by He atoms. This suggests a strong bond between Cs and Pt atoms and the localization of electron cloud density for the (2x2) structure as a result of the charge transfer from Cs atoms to the substrate Pt(111) surface. It is also suggested that the work function minimum, attained at the Cs coverage 0.25 forming the (2x2) structure, is closely related to this charge transfer. On the contrary, the small corrugation for the (√3x√3) structure at a full coverage indicates that a bond among adsorbed Cs atoms is metallic, the electrostatic potential being delocalized along the direction parallel to the surface.
SS2-MoP-17 ARPES Study of 2x2 Mn-Pt(111) Surface Alloys
F.J. Palomares, P.P. Martin, H. Isern, M. Alonso, F. Soria (Instituto de Ciencia de Materiales de Madrid (CSIC), Spain)
The annealing of metallic layers grown on single crystal substrates has enabled the synthesis of ordered surface alloys and metastable phases non existing in the bulk diagram. This behavior has been reported for the Mn-Pt(111) interface, for which two apparently different 2x2 surface ordered alloys are identified as temperature increases. Angle-resolved photoelectron spectroscopy with synchrotron radiation has been used to study the electronic structure of thin films of these two alloys. Normal and off-normal emission geometries are investigated, for photon energies in the range of 12 to 70 eV. The corresponding results are analyzed in comparison with those from the starting Pt(111)-1x1 surface. Our data confirm that such alloy phases exhibit distinct electronic features, which are discussed taking into account the properties calculated for related systems.
SS2-MoP-18 Kinematic Analysis of a Dispersion Region Probed in High-resolution Electron Energy-loss Spectroscopy
T. Inaoka (Iwate University, Japan)
In high-resolution electron energy-loss spectroscopy (HREELS) with the specular-reflection geometry, a probed dispersion region determined kinematically depends upon the incident angle α and the incident energy E0. With change in α and E0, we make a kinematic analysis of the probed dispersion region with special attention to its width. Simultaneously, we evaluate the effect of the image force and the analyzer aperture on the probed region. Our analysis shows that, with a grazing angle α fixed and with E0 varied in a broad range, we can scan a sharply narrowed probed region on the wavenumber-energy (Q-ω) plane to make a close observation of the dispersion relation of surface-excitation modes. Even at grazing incidence, the image force has no substantial influence on the probed region, though it operates to reduce the energy-loss intensity remarkably at low E0. When the parameter hω/2E0 becomes appreciable compared with the aperture angle, presence of the aperture causes a certain decrease in the width of the probed Q region. In the HREELS with the off-specular geometry, the incident energy E0 is restricted to a lower range to insure sufficient intensity, and the probed dispersion region depends upon α and an off-specular angle θ. With change in α and θ, we make the same analysis of the probed dispersion region with close attention to its width as in the specular-reflection geometry. The results of this analysis for the off-specular geometry will also be reported at the conference.
SS2-MoP-19 Spectroscopic Ellipsometry in the Infrared Range Applied on HF-cleaning
J.-C. Cigal, G.M.W. Kroesen (Eindhoven University of Technology, The Netherlands)
Spectroscopic ellipsometry is a powerful non-destructive diagnostic tool for the study of interfaces and layered systems. By combining this method with Fourier transform infrared spectroscopy one can obtain information on the physical parameters of a complex surface like the complex refractive index, and the thickness of the different layers of the system, and on its chemical composition. We report the design of an ellipsometer running on the principle of rotating compensator ellipsometry. This technique offers many advantages compared to other photometric methods. Among them are the non-ambiguous determination of the ellipsometric parameters and the self-calibration of the apparatus. The imperfect components of the different optical elements are taken into account in the calculation algorithms. To illustrate the performance of such an ellipsometer we will clean a silicon wafer with hydrogen fluoride. The surface will be H-passivated. The wafer will be then exposed to open air. The goal of this experiment is to demonstrate the sensitivity of the instrument by analyzing the first steps of the oxidation of the silicon surface.
SS2-MoP-20 Surface Structure Influence on Reactivity of Small Molecules on SrTiO3(100) Surfaces
L. Wang, K.F. Ferris, M.H. Engelhard (Pacific Northwest National Laboratory)
Interactions of water, methanol and HCOOH with stoichiometric, stepped, and reduced SrTiO3(100) surfaces have been studied using temperature programmed desorption (TPD), X-ray photoelectron spectroscopy (XPS), and electronic structural calculations. Comparison of TPD spectra and equilibrium adsorption geometries for H2O on SrTiO3(100) with unreconstructed TiO2 (110) and (100) surfaces reveals the structural influence on the water adsorption and desorption behavior. A unique geometric arrangement of surface atoms on stoichiometric (TiO2-terminated) SrTiO3(100) surfaces due to the lack of surface bridging oxygen atoms results in very different adsorption and desorption properties of H2O on SrTiO3(100) compared to TiO2 (110) and (100) surfaces. Electronic structure calculations show that stepped SrTiO3(100) surfaces enhance the localized interaction between the adsorbed molecule and the oxide surface. No reaction products were observed at above 400 K for methanol on annealed stepped SrTiO3(100). However, reactivity of methanol on SrTiO3(100) surfaces are enhanced for the reducing SrTiO3(100) surfaces prepared by Ar+ sputtering. Reaction products of H2, CH4 and CO were observed at above 400 K on the reduced SrTiO3(100) surface. Formic acid was dissociated to form formate and a surface proton below 250 K on SrTiO3(100) surfaces. Formate was decomposed primarily through dehydration to produce CO and H2O, instead of through dehydrogenation to produce CO2 and H2. Different reaction pathways were observed for formaldehyde formation from formic acid on SrTiO3(100) surfaces. On stoichiometric and stepped surfaces, formaldehyde was produced through bimolecular coupling of two formates. However, on reduced surfaces, formaldehyde formation involves the reduction of surface formate by the oxidation of reduced Ti cations.
SS2-MoP-21 Anomalous Growth of Ba on Ag(111)
O.M.N.D. Teodoro (New University of Lisbon, Portugal); J. Los (FOM- Institute for Atomic and Molecular Physics, NL); A.M.C. Moutinho (New University of Lisbon, Portugal)
The deposition of alkalis or earth alkali elements on other metals often produce a well-known decrease of the surface's work function. Such films make possible to achieve work function values lower than 2 eV. These surfaces found applications, for example, in long life cathodes or in negative ion converters. When these electropositive elements are adsorbed the work function drops steeply and reaches a minimum at coverage lower than 1 monolayer. Then, it increases slightly and the work function converges to the value of the deposited element. In this work, we report an anomalous behavior found during the deposition of barium on an Ag(111) surface. After the minimum of about 2.4 eV the work function did not increase up to 2.7 eV, the bulk barium work function, no matter the amount of barium deposited. These results were corroborated by Auger Electron Spectroscopy where we measured a permanent and constant intensity of the Ag MNN peak for high coverages of barium. To explain this anomalous growth of barium on Ag(111) we propose a model based in the diffusion of silver atoms to the barium film. Details of this model will be presented. Further experiments, showed that co-adsorption of oxygen before a second deposition of barium blocked the diffusion allowing the work function to reach the value of 2.7 eV.
Time Period MoP Sessions | Topic SS Sessions | Time Periods | Topics | AVS2001 Schedule