AVS1997 Session SS1-WeM: Electronic Effects in Structural Studies
Wednesday, October 22, 1997 8:20 AM in Room A1/2-A
Wednesday Morning
Time Period WeM Sessions | Abstract Timeline | Topic SS Sessions | Time Periods | Topics | AVS1997 Schedule
Start | Invited? | Item |
---|---|---|
8:20 AM | Invited |
SS1-WeM-1 Welch Award Invited Talk
Ph. Avouris (IBM T.J. Watson Research Center) |
9:00 AM |
SS1-WeM-3 Atomic and Electronic Structures of Ge(111)-c(2x8) Surface with Defects Studied by STM
G. Lee, H. Mai, I. Chizhov, R.F. Willis (The Pennsylvania State University) Atomic and electronic structures of the sputter-annealed Ge(111)-c(2x8) surface have been studied using scanning tunneling microscopy (STM). The images of both clean surface and the one with defects manifest voltage-dependent variations. While only adatoms of Ge(111)-c(2x8) surface appear with negative sample biases (occupied states), both adatoms and rest-atoms are simultaneously imaged with positive biases (unoccupied states) but changing contrast with varying voltages. The defects present on the surface also exhibited significant voltage-dependent variations in brightness. In particular, delocalized brightness variation was observed around them when bias voltages lower than 1 eV were used. These defects were found to be charged relative to the clean, perfectly-ordered part of the surface, giving rise to the delocalization in the images. We identify these defects and characterize their charged nature. Information about the electronic structure of the Ge(111)-c(2x8) surface will be extracted from the voltage-dependent variation in the images. |
|
9:20 AM |
SS1-WeM-4 Scanning Tunneling Microscopy (STM) and Low Energy Electron Diffraction (LEED) Study of Para- and Meta-Xylene and Its Coadsorption with Carbon Monoxide on Rh(111)
P. Cernota, H.A. Yoon (University of California, Berkeley); M. Salmeron (Lawrence Berkeley National Laboratory); G.A. Somorjai (University of California, Berkeley) STM and LEED have been used to study the surface structures of xylenes on Rh(111). The para- and meta- isomers were used in this study. At high exposures, both xylenes can be imaged at steps on the metal surface, while at low exposures, the molecules are only imaged at the step edges. The two xylenes have different shapes in the STM images. When adsorbed together, the two isomers can be identified by their different shapes, and they exhibit complete mixing. When CO is coadsorbed with each xylene isomer, the LEED pattern shows a new surface structure, and the STM images show that the molecules pack less densely on the surface. The CO is not imaged since its corrugation on the rhodium surface is much smaller than the xylene corrugation. |
|
9:40 AM |
SS1-WeM-5 Cross-Sectional Scanning Tunneling Microscopy at Low Bias
N.D. Jager, D.F. Ogletree (Lawrence Berkeley National Laboratory); E.R. Weber (University of California, Berkeley); M. Salmeron (Lawrence Berkeley National Laboratory) In cross-sectional scanning tunneling microscopy (XSTM) studies of the (110) surface of III-V semi-conductors it is commonly observed that for bias voltages > 2 V, depending on the polarity, either the group III or group V sublattice is imaged. The group III sublattice observed at positive sample polarity relates to empty cation-derived conduction band states, while filled anion-dominated valence band states are imaged at negative sample bias. Therefore simultaneously recorded images at opposite polarity are in different registry, i.e. the single corrugation maximum per unit cell is shifted representing the different geometric positions of the cation and anion, respectively. Our observations on n-GaAs (110) demonstrate that at small bias voltages (< 1V) a polarity change does not alter the image, i.e. only one registry is observed. By taking tip induced band-bending into account we can identify surface conduction band states near the X-point of the surface Brillouin zone to dominate this contrast. The unusual distinct corrugation noticed along [1-10] is in agreement with the symmetry of X. For p-GaAs we can observe two corrugation maxima per unit cell for both polarities at low voltage, i.e. both sublattices are imaged simultaneously. The origin of this new phenomenon is discussed on the basis of details of the valence band structure, tip induced states, or substrate reconstruction. |
|
10:00 AM |
SS1-WeM-6 Scanning Tunneling Microscopy of Wurtzite II-VI Semiconductor Cleavage Surfaces
B. Siemens, C. Domke, Ph. Ebert, K. Urban (Forschungszentrum Jülich GmbH, Germany) Compound semiconductors grown in the hexagonal wurtzite structure attracted growing attention, because many promising materials for blue lasers occur only in the wurtzite structure. Although cross-sectional scanning tunneling microscopy has recently evolved into a powerful tool to study heterostructures of cubic compound semiconductors, the applicability of this particularly fruitful technique on wurtzite compound semiconductors has not been evaluated yet. Therefore, we investigated wurtzite CdSe and CdS (11-20) and (10-10) cleavage surfaces for the first time with atomically resolved scanning tunneling microscopy (STM). The STM images confirm a 1x1 reconstruction coupled with a charge transfer from cations to anions. Thus the occupied and empty state images show the anion and cation derived dangling bonds in the valence and conduction band, respectively. No states in the band gap were found. The electronic structure of the surface permits the observation of dopant atoms in subsurface layers. Even for high step densities the dopant atoms can still be observed, because step edges are facets of neighboring cleavage planes and are uncharged. Thus no band bending obscures the charged point defects and dopant atoms close to steps. We conclude that nonpolar cleavage planes of wurtzite compound semiconductors are well suited for cross-sectional STM studies. |
|
10:20 AM |
SS1-WeM-7 Adsorption and Incorporation of Potassium at the Pt(111) Surface
H. Ibach, A. Grossmann (Forschungszentrum Jülich, Germany); J.B. Hannon (Sandia National Laboratories); C. Klünker, M. Giesen, G. Schulze Icking-Konert (Forschungszentrum Jülich, Germany) The adsorption of K on Pt(111) has been studied quite extensively in the past and was assumed to be well understood in terms of the classical Gurney model for alkali adsorption. We have discovered that K exists in two very different states on Pt(111). In the adsorbed state, the K atom resides in the threefold hollow site. In the second state, K is embedded in the Pt matrix and replaces a second layer Pt atom. The two different states are characterized by an opposite sign of the induced surface stress. The stress induced by adsorbed K is tensile, as expected for a strong electron donor. The induced stress for embedded K is compressive. Electron energy loss spectrocopy (EELS) displays an intense peak at 136 cm-1 for the low coverage adsorbed state. The peak corresponds to perpendicular vibrational motion of the K atom. The frequency shifts to higher value (180 cm-1) and decreases in intensity as the coverage increases, indicating a dramatic reduction of the dipole moment with increasing coverage. Beyond a coverage of 0.1 per surface Pt atom, the adsorbed state transforms gradually into the embedded state when the surface is held slightly above room temperature. The embedded state is characterized by a vibrational frequency of 225 cm-1. Scanning tunneling microscopy (STM) with atomic resolution displays the Pt(111) surface with the embedded K atoms as an intact (111) surface. The presence of the K atom in the second layer underneath the Pt surface layer is indicated by an outwards buckling of Pt atoms. The outwards buckling as well as the compressive stress is assumed to be caused by the larger atom diameter of the embedded K atom compared to Pt. Because of the low frequency parallel vibrations and the strong dependence of the binding energy on the coverage for the adsorbed state, the balance between the two states depends critically on the temperature and the coverage. |
|
10:40 AM |
SS1-WeM-8 The Structure of Buried Atoms in the High Coverage S/Cu(001) System by Chemical-State Photoelectron Diffraction
X. Chen (University of Wisconsin, Milwaukee); J.D. Denlinger (University of Michigan); B.P. Tonner (University of Wisconsin, Milwaukee) The structure of S atoms on Cu(001) at low coverages is one of the classic problems in surface science, and is by now very well understood. At higher coverages, however, the surface forms a complex √(17) x √(17) R 14 structure, which is very stable and occurs in natural circumstances, but has not yet been solved to complete agreement in the literature. A LEED study found a structure in which the high coverage phase consists of 8 sulfur atoms per unit cell, in a single layer at the surface1. A later STM study instead proposed a bilayer structure with sulfur atoms in two layers2. Since the correct structure requires a knowledge of atoms below the surface, which are invisible in the STM images, we have applied the new technique of core level chemical-shift photoelectron diffraction to this problem. Using very high energy and angular resolution x-ray photoelectron diffraction, we have separately measured the diffraction patterns of surface and sub-surface sulfur atoms, by both angle-dependent and energy dependent XPD. The experiments are modelled by single and multiple-scattering theory. We find that the single layer model is not compatible with our results. R-factor analysis of theory and experiment are used to determine the atomic coordinates in the two layer model.
|
|
11:00 AM |
SS1-WeM-9 The Local Adsorption Structure of SO2 on Ni(111): A Normal Incidence X-ray Standing Wavefield Determination
G.J. Jackson, J. Lüedecke, S.M. Driver, D.P. Woodruff (University of Warwick, United Kingdom); R.G. Jones, A. Chan (University of Nottingham, United Kingdom); B.C.C. Cowie (Daresbury Laboratory, United Kindgom) A normal incidence X-ray standing wavefield study of the structure of molecular SO2 on Ni(111) has been conducted, using photoabsorption at both the O and S atoms and real space site triangulation using {111} scatterer planes both parallel to, and at 70° to, the surface plane. Both O and S atoms are found to be in the vicinity of atop sites, although the S atoms are displaced significantly further from these high symmetry sites. S K-edge NEXAFS confirms an earlier finding that the molecule lies with its molecular plane essentially parallel to the surface. The detailed adsorption sites of the O and N species implied by the data analysis can only be reconciled with a model in which the SO2 molecules are centred close to hollow sites (with equal occupation of both types of hollow) and the internal conformation of the molecule, especially the O-S-O bond angle, is significantly different from that of the gas-phase molecule. Specifically, the O-S-O bond angle is estimated to be no more than 100°, while the data indicate an S-O bondlength expansion of 5% or more. This change is attributed to the unusual π-bonding (for which there appears to be no analogue in coordination compounds) and thus partial occupation of the 3b1 π* LUMO of the molecule. |
|
11:20 AM |
SS1-WeM-10 Location of Light atoms by X-ray Standing Waves: Structure of the O/Na/Si(111) Interface
C. Sánchez-Hanke (HASYLAB at DESY, Germany); V. Eteläniemi (University of Helsinki, Finland); A. Hille, G. Materlik (HASYLAB at DESY, Germany); E.G. Michel (Universidad Autonoma de Madrid, Spain) We have investigated the atomic structure of O/Na/Si(111) by using the x-ray standing wave (XSW) technique for alkali metal coverages ≤ 1 ML. Na Kα fluorescence and O 1s photoemission intensity were simultaneously monitored while sweeping over the (111) substrate Bragg reflection. The photon energy was reduced down to approx. 4 keV to increase the intensity of adsorbate signals. The modulation of Na and O signals was used to determine the location of each species with respect to (111) lattice planes for oxygen exposures below 0.5 L. This procedure allowed us to investigate the position of light atoms (as Na and O) using the XSW technique. Na atoms occupy on top sites of the Si(111) clean surface. No adsorption on adatom sites was found. The location of Na atoms is only slightly modified upon oxygen adsorption. Two different O 1s components were resolved, even at the high kinetic energies used (around 3.5 keV). The location of each type of oxygen species has been independently determined by XSW along the oxygen exposure process. We have found that each species presents a different geometric location. The first one keeps approximately the same position as in the clean Si(111) surface for an equivalent exposure (i.e. a bridge geometry). The second one lies at a lower level and most probably corresponds to a bridge geometry modified by the presence of AM atoms. A detailed model for each site will be exposed. |