My research falls under the broad category of ``Condensed Matter Physics,'' that is, I am trying to understand the structural, electronic, and optical properties of solids and, in some cases, liquids. While we can learn about material properties by measuring the response of a sample to external pressure, injection of current, irradiation with light, etc., we also need calculations to help interpret and understand existing experiments and to suggest new ones. This theoretical approach is my main interest and it requires analytical calculations of simplified models and intensive numerical computation of realistic materials.

Besides fundamental properties of matter, some of which I will discuss in the ``Exotic Materials'' section, I am particularly interested in a field called ``Computational Materials Science.'' The main idea behind this buzz word is that the computational power of modern supercomputers combined with constant algorithmic improvements will allow us in the next, say, 10 years to design industrially relevant materials starting from atomistic simulations. In simple terms, we want to write codes for the semiconductor industry that help design chips. The closest analogue are perhaps fluid-dynamics codes in the auto and airplane industry, which have significantly reduced the need for costly wind tunnel experiments and, correspondingly, the time-to-market of new models.

A few examples of ``Computational Materials Science'' are summarized in the section ``Materials Science.'' To see what the future holds, you might want to visit the web site of the Department of Energy ``Center for Computational Materials Simulation'' (CCMS) or of the ``Center of Accelerated Materials Modeling'' (CAMM) at Ohio State University. While there is much more exciting research out there, these web sites will get you started.

Obviously, computational materials science does require numerical computation. A significant part of my work is spent writing, debugging, and optimizing codes. Among other things, I was among the first to adapt object-oriented code design in C++ and parallel computing to computational many-body theory. If you would like to find out more about this, please visit the ``Computation'' section.

After all this applied stuff, you might wonder if I am doing any ``fundamental'' work at all. Indeed, I am particularly interested in improvements and further developments of the so-called ``Density Functional Theory'' and you can read about my work in the corresponding section. Density functional theory was developed in the mid-nineteen sixties by Hohenberg, Kohn, and Sham (my former postdoc advisor). In 1999, Walter Kohn shared the Nobel Prize in Chemistry for his contribution to the development of density functional theory.

Below you find an overview over my projects arranged in three major categories. A pictorial overview that also introduces some of my collaborators is given in this figure.

Density Functional Theory

Basics of Density Functional Theory

Density Polarization Functional Theory

Exact-Exchange Density Functional Theory

and Quasiparticle Calculations

Materials Science of Semiconductors

Structural Properties

Diffusion in Liquid Germanium

Defects in Silicon

Electronic Properties

Review of Quasiparticle Calculations

Valence band off-set in GaN/AlN systems

Optical Properties

General

Electro-Optic Modulation in SiGe

Exotic Materials

Skutterudites

Superconductivity

Computation

Parallel and Object-Oriented Computing for

Computational Many-Body Theory

Platforms