Turbulence and fusion energy rank among the 10 great unsolved problems in physics, according to a recent survey of scientists for the millennium issue of the magazine Physics World . Richard Feynman (1918-1988) believed that comprehending turbulence is the most important challenge of classical physics; after all, it is the usual state of motion for fluids, except at high viscosities.

A modern way to tackle the problem of turbulence is to make use of numerical experiments on powerful supercomputers. These simulations get more demanding and expensive as the temperature of the fluid increases: An evaporated fluid has to be treated as a compressible gas, and, heated further, the gas becomes ionized, becoming a "plasma" of positively and negatively charged particles that interact collectively via electric and magnetic fields. This fourth state of matter constitutes nearly all of the visible universe; plasmas are what stars and our sun are made of. And it is plasmas that we have to deal with when we seek to use the energy source of the sun in fusion power plants on Earth.

I had my first encounter with plasma theory when the subject was taught as an advanced physics course by Professor Dieter Pfirsch during my undergraduate studies at the Technical University of Munich. Then as now, my main interest was mathematical and computational physics. The theory of plasmas was, for me, a particularly attractive field because of the interplay of mechanics, electrodynamics, statistics, and thermodynamics, and also because of its relevance to astrophysics and many laboratory applications. I did not know back then that the professor, who was affiliated with the Max Planck Institute for Plasma Physics (IPP) in Garching, is an eminent figure and one of the pioneers in the field. With this inspiration, I subsequently attended the annual holiday course for European students of physics and graduates at the IPP Summer University for Plasma Physics.

During the semester break of the same year, I worked for several weeks as a student intern in the IPP's technology division. The university offered a plasma seminar, which I took. During the course I contacted Professor Jürgen Meyer-ter-Vehn of the Max Planck Institute for Quantum Optics, where the laser plasma theory group offered me the chance to do research for my diploma thesis on the interaction of strong laser pulses with solid matter.

After graduation from university I joined IPP again to work on my doctoral thesis. The research at IPP's Garching, Greifswald, and Berlin branches concerns the physical basis of a fusion power plant. Like the sun, such a plant is supposed to derive energy from the fusion of atomic nuclei.

In Garching, IPP has two large-scale experiments: the Axially Symmetric Divertor EXperiment (ASDEX) Upgrade tokamak and the WENDELSTEIN 7-AS stellarator. The follow-up stellarator experiment, WENDELSTEIN 7-X, is under construction at the Greifswald branch. Twelve scientific divisions are investigating confinement of high-temperature hydrogen plasmas in magnetic fields, heating of plasmas, plasma diagnostics, magnetic field technology, data acquisition and processing, plasma theory, materials research, plasma-wall interaction, and systems studies. Together with the University of Greifswald, IPP runs the Max Planck International Research School on "Bounded Plasmas," an interdisciplinary graduate school that combines plasma physics with fusion research, computational physics, and surface science.

Nuclear fusion is an important natural process: Many chemical elements originate from hydrogen through fusion; fusion is the energy source of the sun and stars. Under laboratory conditions it is the two hydrogen isotopes--deuterium and tritium--that fuse most readily when held as a plasma at temperatures of several hundred million degrees. In the process a helium nucleus is produced, accompanied by release of a neutron and large quantities of usable energy. Fusion fuels are cheap and widely available on Earth. Moreover, because a fusion power plant would have ecologically favorable properties, fusion could make an enduring contribution to our future energy supply.

Because of its high temperature, a fusion plasma cannot be directly confined in material vessels. Any wall contact would immediately recool the thin gas. The problem is solved by using magnetic fields, which confine and thermally insulate the charged particles in the fuel, keeping them away from material surfaces.

Consequently, fusion plasmas possess large differences in temperature and density between the hot, dense center (where heating and refueling is applied) and the cooler wall area (where plasma may leak out). The whole system is being constantly driven and is always very far from thermodynamic equilibrium. As a consequence, instabilities in the plasma motion and magnetic field occur. Identifying their causes and finding remedies has been one of the main fields of endeavor, particularly in the early days of fusion research. In most cases, it required a lengthy series of experiments and intensive collaboration between experimental and theoretical physicists. One of the most harmful phenomena these investigations have discovered is the drift instability, which leads to small-scale turbulence of the plasma that efficiently transports heat and particles by convection to the outer regions, where they are lost and unable to contribute to nuclear fusion.

For my doctoral thesis I investigated effects of the particular magnetic field shape from different experiments, such as WENDELSTEIN 7-X, on this drift instability and turbulence. For these comparative studies I applied both simple model calculations and massive three-dimensional direct numerical simulations on supercomputers.

In the end it turned out that some configurations seem to be more favorable than others, and turbulent transport losses may be reduced by clever optimization of the magnetic field, hence increasing the efficiency of a prospective power plant.

However, one quickly reaches the bounds of today's available computing power when large-scale experiments need to be modeled numerically while retaining all important physical effects, and exhaustive simulations of fusion reactor plasmas as a whole are still not feasible. But besides having applications to possible power plants, there is still also a lot to investigate in the basic physics of turbulence in 3D inhomogeneous magnetized plasmas. Subtle nonlinear effects that change the properties of drift instabilities in fully developed turbulence make life even more interesting. As a postdoc in the tokamak turbulence theory group, I am currently involved in further developing our numerical methods, and I continue my work on the influence of magnetic field shape in tokamaks such as ASDEX Upgrade.

Young scientists in fusion research are watching closely the ratification of the joint implementation agreement on the International Thermonuclear Experimental Reactor (ITER) as the next major step for the development of fusion. A decision is due within the next 2 years.

Further detailed information on degree and Ph.D. thesis work opportunities as well as open postdoc positions at IPP are available at the institute's Web page.