Novel genetic algorithm search procedure for LEED surface structure determination.

Low Energy Electron Diffraction (LEED) is one of the most powerful experimental techniques for surface structure analysis but until now only a trial-and-error approach has been successful. So far, fitting procedures developed to optimize structural and nonstructural parameters-by minimization of the R-factor-have had a fairly small convergence radius, suitable only for local optimization. However, the identification of the global minimum among the several local minima is essential for complex surface structures. Global optimization methods have been applied to LEED structure determination, but they still require starting from structures that are relatively close to the correct one, in order to find the final structure. For complex systems, the number of trial structures and the resulting computation time increase so rapidly that the task of finding the correct model becomes impractical using the present methodologies. In this work we propose a new search method, based on Genetic Algorithms, which is able to determine the correct structural model starting from completely random structures. This method-called here NGA-LEED for Novel Genetic Algorithm for LEED-utilizes bond lengths and symmetry criteria to select reasonable trial structures before performing LEED calculations. This allows a reduction of the parameter space and, consequently of the calculation time, by several orders of magnitude. A refinement of the parameters by least squares fit of simulated annealing is performed only at some intermediate stages and in the final step. The method was successfully tested for two systems, Ag(1 1 1)(4 × 4)-O and Au(1 1 0)-(1 × 2), both in theory versus theory and in theory versus experiment comparisons. Details of the implementation as well as the results for these two systems are presented.

Optimal electron doping of a C60 monolayer on Cu(111) via interface reconstruction.

We demonstrate the charge state of C60 on a Cu(111) surface can be made optimal, i.e., forming C60(3-) as required for superconductivity in bulk alkali-doped C60, purely through interface reconstruction rather than with foreign dopants. We link the origin of the C60(3-) charge state to a reconstructed interface with ordered (4x4) 7-atom vacancy holes in the surface. In contrast, C60 adsorbed on unreconstructed Cu(111) receives a much smaller amount of electrons. Our results illustrate a definitive interface effect that affects the electronic properties of molecule-electrode contact.

Stacking of polycyclic aromatic hydrocarbons as prototype for graphene multilayers, studied using density functional theory augmented with a dispersion term.

The interlayer pi-pi interaction between finite-size models of graphene sheets was investigated by using a density functional theory method, augmented with an empirical R(-6) term for the description of long-range dispersive interaction; these were calibrated by studying the pi-pi interaction between various benzene dimer configurations and comparing the results with previous calculations. For stacked bilayers (dimers) and multilayers of polyaromatic hydrocarbons, which serve as molecular models of graphene sheets, we found that binding energies and energy gaps are strongly dependent on their sizes, while the stacking order and the number of stacked layers have a minor influence. The remarkably broad variation of the energy gap, ranging from 1.0 to 2.5 eV, due mainly to variation of the model size, suggests the potential of broadband luminescence in the visible range for carbon-based nanomaterials that have pi-pi interacting.

Efficient calculation of electron diffraction for the structural determination of nanomaterials.

A critical advance in the technique of low-energy electron diffraction is presented and shown to enable determining detailed structures of nanomaterials, based on experimental methods that already exist or have been proposed. Our new cluster approach speeds up the computation to scale as n logn, rather than the current n3 or n2, with n the number of atoms, for example. Applications are illustrated for C60 molecules adsorbed on a Cu(111) surface, with and without coadsorbed metal atoms, exhibiting sensitivity to important structural features such as buckyball size and deformation.

Monte Carlo simulations of segregation in Pt-Ni catalyst nanoparticles.

We have investigated the segregation of Pt atoms in the surfaces of Pt-Ni nanoparticles, using modified embedded atom method potentials and the Monte Carlo method. The nanoparticles are constructed with disordered fcc configurations at two fixed overall concentrations (50 at. % Pt and 75 at. % Pt). We use octahedral and cubo-octahedral nanoparticles terminated by {111} and {100} facets to examine the extent of the Pt segregation to the nanoparticle surfaces at T=600 K. The model particles contain between 586 and 4033 atoms (particle size ranging from 2.5 to 5 nm). Our results imply that a complete {100}-facet reconstruction could make the cubo-octahendral Pt-Ni nanoparticles most energetically favorable. We predict that at 600 K due to segregation the equilibrium cubo-octahedral Pt50Ni50 nanoparticles with fewer than 1289 atoms and Pt75Ni25 nanoparticles with fewer than 4033 atoms would achieve a surface-sandwich structure, in which the Pt atoms are enriched in the outermost and third atomic shells while the Ni atoms are enriched in the second atomic shell. We also find that, due to an order-disorder transition, the Pt50Ni50 cubo-octahedral nanoparticles containing more than 2406 atoms would form a core-shell structure with a Pt-enriched surface and a Pt-deficient homogenous core.

Surface structures of cubo-octahedral Pt-Mo catalyst nanoparticles from Monte Carlo simulations.

The surface structures of cubo-octahedral Pt-Mo nanoparticles have been investigated using the Monte Carlo method and modified embedded atom method potentials that we developed for Pt-Mo alloys. The cubo-octahedral Pt-Mo nanoparticles are constructed with disordered fcc configurations, with sizes from 2.5 to 5.0 nm, and with Pt concentrations from 60 to 90 atom %. The equilibrium Pt-Mo nanoparticle configurations were generated through Monte Carlo simulations allowing both atomic displacements and element exchanges at 600 K. We predict that the Pt atoms weakly segregate to the surfaces of such nanoparticles. The Pt concentrations in the surface are calculated to be 5-14 atom % higher than the Pt concentrations of the nanoparticles. Moreover, the Pt atoms preferentially segregate to the facet sites of the surface, while the Pt and Mo atoms tend to alternate along the edges and vertexes of these nanoparticles. We found that decreasing the size or increasing the Pt concentration leads to higher Pt concentrations but fewer Pt-Mo pairs in the Pt-Mo nanoparticle surfaces.

Monte Carlo simulations of segregation in Pt-Re catalyst nanoparticles.

We have investigated the segregation of Pt atoms to the surfaces of Pt-Re nanoparticles using the Monte Carlo method and modified embedded-atom method potentials that we have developed for Pt-Re alloys. The Pt(75)Re(25) nanoparticles (containing from 586 to 4,033 atoms) are assumed to have disordered fcc configurations and cubo-octahedral shapes (terminated by [111] and [100] facets), while the Pt(50)Re(50) and Pt(25)Re(75) nanoparticles (containing from 587 to 4,061 atoms) are assumed to have disordered hcp configurations and truncated hexagonal bipyramidal shapes (terminated by [0001] and [1011] facets). We predict that due to the segregation process the equilibrium Pt-Re nanoparticles would achieve a core-shell structure, with a Pt-enriched shell surrounding a Pt-deficient core. For fcc cubo-octahedral Pt(75)Re(25) nanoparticles, the shells consist of almost 100 at. % of Pt atoms. Even in the shells of hcp truncated hexagonal bipyramidal Pt(50)Re(50) nanoparticles, the concentrations of Pt atoms exceed 85 at. % (35 at. % higher than the overall concentration of Pt atoms in these nanoparticles). Most prominently, all Pt atoms will segregate to the surfaces in the hcp truncated hexagonal bipyramidal Pt(25)Re(75) nanoparticles containing less than 1000 atoms. We also find that the Pt atoms segregate preferentially to the vertex sites, less to edge sites, and least to facet sites on the shell of Pt-Re nanoparticles.

The NIST Surface Structure Database -- SSD version 4.

SSD-4, the fourth version of the NIST Surface Structure Database, has appeared in early 2002 as a significant upgrade to both the database and graphics software; it now also includes nearly 1300 structures published through the end of 2000. SSD is an interactive PC-based database of critically selected surface structures determined with a variety of experimental techniques. The data include not only atomic coordinates, bond lengths and bond angles, but also information about experimental preparation, measurement and analysis methods. The software provides advanced search and display facilities, as well as interactive three-dimensional color visualization and analysis tools of great flexibility. In addition, the program can generate publication-quality color or grayscale prints of any structure, with a host of user-selectable options (such as view angle, perspective, colors, ball and bond styles, and labels).

Differential photoelectron holography: a new approach for three-dimensional atomic imaging.

We propose differential holography as a method to overcome the long-standing forward-scattering problem in photoelectron holography and related techniques for the three-dimensional imaging of atoms. Atomic images reconstructed from experimental and theoretical Cu 3p holograms from Cu(001) demonstrate that this method suppresses strong forward-scattering effects so as to yield more accurate three-dimensional images of side- and backscattering atoms.

Circular dichroism in K-shell ionization from fixed-in-space CO and N2 molecules.

We have measured the angular distributions of 1s photoelectrons excited by circularly and linearly polarized light from fixed-in-space CO and N2 molecules, in the vicinity of their shape resonances. A strong circular dichroism, i.e., a strong dependence on the sense of rotation of the polarization vector of the photons, is found for both molecules. State-of-the-art one-electron multiple scattering and partially correlated random phase approximation calculations are in good agreement with many, but not all, aspects of the experimental data.