We report the results of self-consistent quantum Monte Carlo simulations of the structure of the liquid-vapor interfaces of Ga and of alloys of Bi in Ga and In in Ga. The single particle density distribution along the normal to the interface is predicted to display stratification, with a spacing of about an atomic diameter, and the pair structure function in the plane of the liquid-vapor interface is predicted to be essentially the same as that in the bulk liquid. In all of the cases considered the qualitative and quantitative character of these predictions are in good agreement with the results of experimental studies.

We have numerically studied the surface tension and surface profiles of several liquid semiconductors, including Si, Ge, GaAs, CdTe, and their alloys, as a function of temperature and concentration. Two kinds of simulations have been carried out: direct free-energy calculations using Monte Carlo methods, and force summations using molecular dynamics. We use empirical two- and three-body interatomic interactions based on the form originally proposed by Stillinger and Weber for Si, in conjunction with simulation cell sizes ranging from 216 to as large as 8000 atoms and several novel numerical techniques including a direct calculation of the surface entropy. In the case of alloys, we find a striking segregation of the low-surface-tension component to the surface even when the alloy components are miscible at all concentrations.

Two new versions of the density functional theory (DFT), which correspond to the canonical ensemble and the Gibbs ensemble respectively, are applied to study multiple equilibrium states and associated hysteresis on examples of capillary condensation of argon in nanopores of MCM-41 mesoporous molecular sieves and Kr adsorption on carbon. The canonical ensemble DFT (CEDFT) allows us to trace not only the stable and the metastable states along the hysteresis loop, but also the unstable states inside the hysteresis loop. The Gibbs ensemble DFT (GEDFT) allows us to determine precisely the points of equilibrium phase transitions in confined fluids.

We report the results of the Monte Carlo simulation study of the fluid phase transitions in micropores. The Gibbs, Canonical, and Grand Canonical ensembles were employed to obtain the adsorption isotherms of the Lennard-Jones nitrogen in carbon slit micropores at 77.4 K. It is shown for the first time that depending on the choice of the parameters of the Monte Carlo simulations, such as the number of particles in the simulation cells, the sampling from the Gibbs ensemble may supply either van der Waals-type or hysteresis loops in the phase transition zones. The existence of these loops is confirmed by independent sampling from the Canonical ensemble using Widom method of test particles. The Grand Canonical ensemble simulations always produce hysteresis loops within the phase transition zone. Outside the phase transition zones, sampling from all three ensembles yield essentially identical isotherms.

Using Molecular Dynamics, the evolution dynamics of Sn on the (111) and (100) surfaces of Cu have been investigated as a function of coverage and temperature. The interaction potentials are described by modified embedded atom method (MEAM) potentials. The calculated diffusion activation energies of Cu in Sn and Sn in Cu agree reasonably well with experimental values. We find that the structure of the overlayer depends on the morphology of the substrate and remains stable up to temperatures of the order of 70% of the melting temperature of the substrate at which diffusion of Sn into the substrate and Cu atoms onto the overlayer is observed.

Scanning tunneling microscopy (STM) is one of the most successful experimental tools for probing the structure of semiconductor surfaces. However, care must be taken in interpreting the images at the atomistic limit. Often a “naive” interpretation of the STM image can yield an incorrect surface structure. We illustrate this situation via ab initio pseudopotential calculations for the STM image of the (110) GaAs surface. We will compare theoretical STM images to experimental images for the relaxed surface and for a surface with an As vacancy.

Using first principles molecular dynamics, we have studied the reconstructions and thermal properties of the (001) surfaces of cubic SiC. Our calculations show that C-terminated surfaces can have different reconstructions, depending on preparation conditions and thermal treatment, and that the Si-terminated surface geometry can be substantially affected by the presence of stress. Our findings allow us to interpret recent experiments about (001) SiC surfaces, in particular STM images.

We use a local-orbital formalism based on density-functional theory to investigate the stoichiometry and structure of the cation- and anion-terminated (0001) surfaces of wurtzite-phase GaN and A1N. We compare the total energies computed for various p(2×2) reconstructions. First-layer atom deficient structures such as vacancies are found to be the most stable configurations for the anion- and cation-terminated surfaces. For metal rich growth conditions our calculations favor the adsorption of metal atoms. Surface chemical reactions relevant for the growth of thin nitride films, such as the adsorption of hydrogen and nitrogen from decomposed ammonia, are discussed. We determine the total energy differences for the co-adsorption of NH2 and H on different surface structures.

The structural and electronic properties of four different structures of zirconia (ZrO2) are studied using ab initio total energy calculations. The calculations are made in the framework of density functional (DFT) and pseudopotential theory. We compare results given within the LDA (Local Density Approximation) and including Generalized Gradient Corrections (GGCs) in the Perdew Wang and Perdew Becke formalisms. We present results for pure and defective zirconia (oxygen vacancies and Zr substitution by Fe) showing the effects of such point defects on tne relative structural stabilities of the different pseudopolymorphs.

The methods for computer simulation of grain boundaries and hetero-interfaces in ceramics have long been available. We review developments in the field in two areas; the study of general grain boundaries and the simulation of hetero-interfaces. In both cases we discuss extrapolation from the symmetric case. We will also consider the validity of the continued use of classical potential models in an age of large-scale quantum calculations.

The adhesion geometries of coherent cube-on-cube interfaces between spinel (MgAl2O4) and the two metals Al and Ag were determined by density functional band structure calculations in the local density approximation (LDA). For Al/MgAl2O4, for which experimental data are available, the calculated optimum interface geometry is in excellent agreement with HRTEM measurements (distance dint: 1.90 Å calc., 1.90±0.04 Å exp.).

The work of adhesion Wad is calculated for three different high-symmetry translation states between an Al-0 terminated (001) surface of the spinel and the (001) surface of each of the metals. The binding energy curves display a distinct optimum for the adhesion of aluminum atoms on top of the spinel oxygen ions at a Wad value of 2.4 J/m2. For silver several adsorption sites are isoenergetic at 1.1 J/m2 and the intersections of the Wad (dint) curves indicate a low-energy dissociation path. A further analysis of the electronic structure in the Al/MgAl2O4 system reveals the charge redistribution in the metal towards the oxygen ions as the main contribution to bonding. On the contrary, polarization of the metal film is the major effect observed on the adhesion of Ag to the spinel substrate.

{222}MgO/Cu is one of the most extensively characterized ceramic/metal interfaces, in view of the atom-probe field-ion-microscopy, Z-contrast scanning-transmission-electron-microscopy (STEM), and spatially-resolved electron-energy-loss-spectroscopy (EELS) measurements performed by the present authors, as well as the high-resolution electron microscopy (HREM) of this system by others. Atomistic simulations with local density functional theory (LDFT) and molecular dynamics (MD) have been performed to gain additional insight into the structure of this interface. This presentation describes an interface interatomic potential for {222}MgO/Cu derived from LDFT total energy calculations, and its application to structural properties, including the terminating species, the absence of dislocation standoff, and the symmetry of the interfacial dislocation network.

The equilibrium structures as well as the electronic Schottky barriers for (100) Erbium-Arsenide/Gallium-Arsenide (ErAs/GaAs) arsenic and gallium terminated interfaces have been determined by ab-initio calculations using the local-density approximation and a full-potential linear-muffin-tin-orbital method. In both cases the arsenic sublattice was chosen to be continuous across the interface in accordance with experiments on Rutherford backscattering channeling. Band structures, densities of states, and charge density distributions were also determined for the interfaces. The comparison of the total supercell energies reveals that the gallium terminated (chain) interface is more energetically stable than the arsenic terminated (shadow) interface. It also shows that the equilibrium interface separation for the arsenic terminated interface corresponds to an ideal structure when arsenic forms undistorted face-centered cubic lattice. The separation in the gallium terminated interface is quite substantial and is 60% larger than that of the ideal situation. The model also predicts that no buckling of the ErAs interface monolayer will occur for either structure. The computed Schottky barriers for holes (after a semi-empirical quasiparticle self-energy correction) are 0.6 eV for the chain interface and 0.4 eV for the shadow interface.

We describe a new method of examining defects and their interactions with each other at a microscopic level of detail. This involves efficiently calculating local properties at arbitrary spatial points, applicable in periodic supercell calculations. In particular, we have identified two quantities, the local energy density, ε(r) and the local stress density, σαβ(r), which are constrained to yield the total energy and the average macroscopic stress tensor, respectively, when integrated over the entire volume of the supercell. In systems with defects, these microscopic quantities provide important insights about the local nature and environment of defects. For the test case of bulk Al with a point defect, we demonstrate that these concepts result in meaningful local quantities characteristic to the point defect. We propose to use these methods to study the microscopic nature of vacancies at a grain boundary and their interaction with each other.

Atomistic simulations of grain-boundary structures in body-centered cubic transition metals have revealed that angle-dependent contributions to interatomic interactions are essential. Unfortunately, the results of presently available empirical many-body potentials are not yet always sufficiently reliable for quantitative theoretical predictions of grain-boundary structures, which are consistent with experimental observations, e.g. by high-resolution transmission electron microscopy.

Ab-initio electronic-structure calculations based on the local-density-functional theory offer the possibility to determine accurately the microscopic structures of special, high-symmetry grain boundaries, which can be used as data bases for the improvement of empirical many-body potentials. Such ab-initio calculations, with a mixed-basis pseudopotential method and grain-boundary supercells, are presented for Σ5 (310) [001] 36.87° symmetrical tilt grain boundaries in Niobium and Molybdenum.

We have examined a variety of structures for the {510} symmetric tilt boundary in Si, using first-principles calculations. These calculations show that the observed structure in Si is the lowest energy structure. This structure is more complicated than what is necessary to preserve four-fold coordination. We compare the results to classical and tight-binding models, in order to test these empirical approaches.

Available experimental data concerning two possible atomic structures (A and B) of the Σ=11 <011> tilt grain boundary in silicon and germanium are analyzed. Previous empirical calculations concerning low and high temperature stability of these structures are summarized and criticized. New calculations using the semi-empirical tight-binding method (SETBM) are presented for low temperature stability. The tight-binding parameters of Mercer and Chou give results in agreement with experimental observations by high resolution electron microscopy.

Following Mandelbrot's fractal theory, the irregular characteristics of the microstructural features of ferroelectric Pb(Sc0.5 Ta0.5)O3 ceramics, including grain boundaries and dislocation networks, were investigated. The microstructural features were imaged by electron microscopy. The fractal analyses were carried out manually and by image processing techniques, which show the value of the fractal dimension, D, varies according to the regularity of the microstructure. The value of, D, close to unity is an indication of an increasing degree of microstructural regularity, which is in good agreement with the simulated results.

The energy differences between various SiC polytypes are calculated using the full-potential linear muffin-tin orbital method and analyzed in terms of the anisotropie next nearest neighbor interaction (ANNNI) model. The fact that J1 + 2J2 < 0 with J1 > 0 implies that twin boundaries in otherwise cubic material are favorable unless twins occur as nearest neighbor layers. Contrary to some other recent calculations we find J1 > |J2|. We discuss the consequences of this for stabilization of cubic SiC in epitaxial growth, including considerations of the island size effects.

We describe a new simulation method for the preparation of porous silica and present results from molecular dynamics simulations of the structures obtained. We start from a homogeneous liquid phase with reduced atomic charges. The charges are then slowly rescaled and the atoms start clustering to finally form a porous network. We observe that local ordering precedes formation of long range correlations. We investigate physical properties of porous silica such as porosity, internal surface and fractality. They are in reasonable agreement with experimental data, although internal surface and porosity seem to be systematically larger than those found in adsorption experiments. The vibrational and dielectric power spectra show an enhanced intensity in the low frequency region. These modes can be associated with slow dynamics of clusters.