US research initiative: IPP is a partner in developing reliable simulations for fusion power plants

The Max Planck Institute for Plasma Physics (IPP) is now a partner in a newly launched funding program of the US Department of Energy (DOE).

January 31, 2025

As part of a project worth a total of 14 million US dollars, the IPP, together with US institutions, will develop reliable simulations for key requirements of nuclear fusion power plants. The GENE-X numerical code developed at the IPP plays a key role in this.

*Prof. Dr. Frank Jenko heads the Tokamak Theory division at the IPP in Garching near Munich. His role as an adjunct professor at the University of Texas at Austin is the link to the US project.*

Photo: MPI for Plasma Physics, Michael Herdlein

*Prof. Dr. Frank Jenko heads the Tokamak Theory division at the IPP in Garching near Munich. His role as an adjunct professor at the University of Texas at Austin is the link to the US project.*

Photo: MPI for Plasma Physics, Michael Herdlein

In January, the US Department of Energy (DOE) named six projects to receive a total of 107 million US dollars as part of theFusion Innovative Research Engine (FIRE) Collaboratives funding initiative. In one of the projects, the Max Planck Institute for Plasma Physics (IPP) will be the only non-American cooperation partner to develop prediction models for the first fusion power plants – together with eleven American research institutions and companies. On the IPP side, Director Frank Jenko and his Tokamak Theory division in Garching will drive the project forward. He is also an adjunct professor at the University of Texas at Austin, which provides a direct link between IPP and the US project.

Predicting the plasma core and edge equally

The main aim of the 14 million US dollar project (title: Advanced Profile Prediction for Fusion Pilot Plant Design) is to reliably predict the behavior of fusion plasmas in the edge areas. “Our numerical models are already very good for the plasma core. But at the edge, however, such complex physical processes take place that we have reached the limits of what we could do with the previous possibilities,” explains Prof. Jenko. ”To estimate the performance of a future power plant, however, we need to know the temperature and density profiles across the entire plasma. Only now do we have the tools for such predictions.” The GENE-X computer code developed at IPP, which has been specially optimised for calculating turbulence at the plasma edge, plays a central role in this. “Recently, we were able to increase the computing speed for plasma simulations by a factor of 500 – and we still see potential for further improvements,” says Prof. Jenko.

New possibilities through a dramatically faster simulation code

Speed is a crucial factor in plasma physics simulations: today's supercomputers often take weeks to simulate plasma turbulence over a few milliseconds. “An efficient and at the same time physically advanced code like GENE-X will thus advance fusion research enormously. We will achieve an additional leap in knowledge by combining GENE-X with methods of artificial intelligence,” explains Prof. Jenko. The aim of the entire project is to support the development of power plants through simulations – for both main concepts of magnetic confinement nuclear fusion: tokamak and stellarator. The project's proximity to practical application is ensured by constant interaction with an advisory board, whose members include representatives of fusion companies and the international experimental reactor ITER in the south of France.

Cross-section of a tokamak plasma: Scientists used the GENE-X code to simulate turbulence in a specific plasma scenario (L-mode) in the ASDEX Upgrade tokamak. The simulation shows the typical behavior of the turbulence, with density fluctuations that were stronger at the edge (“ballooning”). The GENE-X code demonstrates its ability to simulate turbulent plasma dynamics with exceptional accuracy: The calculated results reproduce experimental observations.

Source: D. Michels et al., Phys. Plasmas 29, 032307 (2022); https://doi.org/10.1063/5.0082413

Cross-section of a tokamak plasma: Scientists used the GENE-X code to simulate turbulence in a specific plasma scenario (L-mode) in the ASDEX Upgrade tokamak. The simulation shows the typical behavior of the turbulence, with density fluctuations that were stronger at the edge (“ballooning”). The GENE-X code demonstrates its ability to simulate turbulent plasma dynamics with exceptional accuracy: The calculated results reproduce experimental observations.

Source: D. Michels et al., Phys. Plasmas 29, 032307 (2022); https://doi.org/10.1063/5.0082413

The project should clarify, for example, how the heat generated in fusion plasmas can be exhausted in a controlled manner. Temperatures of more than 100 million degrees Celsius arise inside, which do indeed decrease to a few thousand degrees at the edge. Nevertheless, the plasma power would damage any wall material. Therefore, strategies are needed to exhaust this power in a controlled manner. The greatest heat load occurs at the so-called divertor, which has the task of extracting contaminations from the plasma. Simulations should help to distribute the heat output over a larger area and to find the optimal design for the divertor.

Modeling of fusion power plants with tungsten wall

The researchers will also investigate how the use of the currently favored wall material tungsten affects the performance of fusion plasmas. Tungsten has the highest melting point of all metals (about 3400 degrees Celsius) and is therefore particularly well suited to withstand the heat loads in the proximity of fusion plasmas. However, when operating in fusion devices, tungsten atoms dissolve and contaminate the plasma. They can cool it significantly and in extreme cases even extinguish it. Prof. Jenko: “To eliminate this undesirable effect, we need to understand the processes involved. This is one of the goals of the new project.”