Roy Richter

Spin anisotropy of ferromagnetic slabs and overlayers (invited)

We calculate the spin anisotropy of ferromagnetic monolayers of Fe, Ni, V, and Co. We find the easy direction of magnetization is perpendicular to the plane of the monolayer for Fe and V, but in the plane for Ni and Co. The result for Fe explains why spin splitting but no spin polarization is observed in recent photoemission experiments on Fe monolayers. For thicker ferromagnetic slabs, the depolarization energy will quickly overwhelm the spin anisotropy and force the moment into the plane of the slab. This is illustrated by calculations for thicker slabs of Fe. Preliminary calculations of the anisotropy of a monolayer of Fe on Ag indicate that the Ag must be treated fully relativistically.

Computer simulations of soot aggregation

Abstract Soot aggregates exhibit a variety of morphologies: long chain-like structures and compact blob-like lumps are two extremes. We have simulated the aggregation of soot spherules in two and three dimensions using a model proposed by Witten and Sander. The individual soot spherules diffuse on a lattice of points before contacting the main aggregate. Refinements to Witten and Sanders model have been implemented: Variable spherule size, sticking probabilities, and surface energy have been included; charged spherules have been allowed to diffuse on the lattice. Analysis by visual means, calculation of bond order, and correlation functions indicate that the physically reasonable models proposed so far do not reproduce some aspects of observed soot aggregates, such as the long chains seen by Stevenson. The correlation functions show the aggregates formed are self-similar—i.e., they have the same appearance on a variety of length scales. Employing the concept of fractal geometry, we find the fractal dimension D of all these aggregates, regardless of size or charge, to be 2.4 ± 0.2. This agrees with similar calculations by Meakin who finds D = 2.51 for our simplest model.

Electronic structure and magnetism of a Pd monolayer on Fe(100)

Abstract Spin-polarized self-consistent localized-orbital (SCLO) calculations have been performed for a five-plane slab simulating Pd/Fe(100). Pd monolayers were placed in registry with both faces of a three-plane Fe(100) slab. We chose a PdFe bond length equal to the sum of metallic radii, 2.62 A, and an FeFe bond length equal to the bulk value, 2.49 A. The computed energy bands for the Pd/Fe 3 /Pd slab resemble those calculated for a five-plane Fe(100) slab, except for a positive work-function shift of 0.5 eV. The Pd monolayer has a magnetic moment of 0.37 μ B /atom. The magnetic moment of an adjacent iron atom is 2.74 μ B , slightly smaller than the value 2.89 μ B at the surface of Fe 5 , but still significantly larger than the central-plane value 2.37 μ B . The d bands of the two metals are strongly hybridized, but very little charge transfer takes place across the interface. Compared with the isolated Pd(100) monolayer, or the clean Fe(100) slab, the surface density of states of Pd/Fe(100) is rather weak near the Fermi level, suggesting a reduced chemical reactivity for this surface.

Component placement in VLSI circuits using a constant pressure Monte Carlo method

Abstract When laying out a VLSI circuit on a silicon wafer the object is to pack the components of the circuit onto a wafer of minimum area subject to a variety of conflicting constraints associated with electrical interconnections among the components and input/ output connections. A constant pressure Monte Carlo method is applied to an idealized component placement problem where the object is to pack different rectangular components onto a square of minimum area with a subsidiary objective of minimizing the total length of wires interconnecting the components. The Monte Carlo method is found to be remarkably effective in solving this idealized problem. No other method for solving this problem is known.

UNIVERSAL PROPERTIES OF BONDING AT METAL INTERFACES

Our knowledge of the fundamental properties of surfaces and interfacial bonds has improved considerably over the last decade. While the fracture process is complex, it is dependent in part on these properties. Thus it is of interest to consider current theoretical understanding of metal surfaces and bimetallic interfacial bonds.

Spin—orbit induced orbital angular momentum in monolayers

Abstract We calculate spin—orbit induced orbital angular momentum in ferromagnetic monolayers of Fe, Co and Ni. For Fe the induced momentum is moderate leading to a spectroscopic splitting factor of 2.06. This is slightly smaller than the value 2.09 measured in bulk Fe. Values in this range are also observed in Fe overlayers on inert substrates. For Co the induced momentum is larger and gives a spectroscopic splitting factor of 2.25, somewhat larger than the bulk Co value of 2.19. For Ni the induced momentum is large: comparable to the spin moment itself. This leads to a prediction of spectroscopic splitting factors as large as 3.06 (for a monolayer with the lattice constant of Ag). Bulk Ni is 2.18. No measurements of the spectroscopic splitting factor for Co or Ni overlayers have been reported.