Introduction
The quantitative description of the structural, electronic, and optical
properties of metals, semiconductors, and insulators is one of the
long-standing challenges of condensed matter physics. Theoretical
methods in combination with computational tools currently provide a
reliable understanding of the properties of bulk materials. More
complicated systems such as extended defects, grain boundaries,
dislocations, etc. are currently under investigation.
At Ohio State, we use high-performance computing on massively
parallel architectures to describe the properties of semiconductors,
insulators, and metals. Examples of this effort can be found under
the links given above.
Structure is generally determined by total energy calculations or by
molecular dynamics calculations. Often, we use a hierarchical approach:
We start investigating a system with classical potentials, which describe
the interaction between atoms by pair potentials, to get a rough idea
of the fundamental processes in and properties of a system. Then we
refine our understanding by using tight-binding. Finally, detailed
questions regarding structure and stability are answered by
first-principles calculations based on density functional theory in the
local density or generalized gradient approximations.
The electronic structure is determined by quasiparticle calculations in
the so-called GW approximation. These calculations allow an accurate
determination of the electronic structure of materials (to within
0.1-0.5 eV). Input for quasiparticle calculations comes either from
LDA or exact-exchange-plus-LDA-correlation density functional
calculations (see also Density Functional Theory).
We have developed parallel codes both for a real-space and a
reciprocal-space GWA calculation.
Last but not least, we determine the linear and nonlinear optical properties
of materials from first-principles. Here, we include - in contrast to
most other groups - local field effects in the calculation. Local field
effects describe screening because of an inhomogeneous density
distribution in a solid and are not negligible. They amount to about
10-15% for the dielectric constant, 10-30% for the coefficient of
second-harmonic generation, and about 100% for the coefficient of
optical activity.
Topics Covered
To cite this page:
Materials Science
<http://www.physics.ohio-state.edu/~aulbur/materials.html>