ERIK MCLEAN / UNSPLASH

Physikalisches Kolloquium

Freitag, 14. November 2025 17:00 Uhr  News from Wendelstein 7-X: Towards Clean Energy from Fusion

Prof. Dr. Thomas Klinger , Max-Planck-Institut für Plasmaphysik, Garching High performance steady state fusion plasmas in the superconducting stellarator device Wendelstein 7-X T. Klinger, Greifswald/Germany Max-Planck Institute for Plasma Physics, Wendelsteinstrasse 1, 117489 Greifswald The stable generation of high-temperature and low-density hydrogen plasmas (ion and electron temperature in the range 10-20 keV resp. 100-200 million degree Kelvin) is the basis for the use of nuclear fusion to generate heat and electric power. The most promising path is to use strong, toroidally shaped, twisted magnetic fields to confine the electrically charged plasma particles in order to avoid heat losses to the cold, solid wall elements. Two magnetic confinement concepts have proven being most suitable: (a) the tokamak and (b) the stellarator. The stellarator creates the magnetic field by external coils only, the tokamak by combining the externally created field with the magnetic field generated by a strong current in the plasma. “Wendelstein 7-X” is the most advanced large superconducting stellarator that operates at the Max Planck Institute in Greifswald. With 30 m3 plasma volume, 3 T magnetic field on axis, and 10 MW micro wave plasma heating power, hydrogen plasmas are generated that allow one to establish a technical and scientific basis for the extrapolation to a future fusion power plant. It is a unique feature of Wendelstein 7-X to be able to operate high-power hydrogen plasmas under steady-state conditions, more specifically for 1800 s (note that the world standard is now in the 10 s ballpark). Furthermore, Wendelstein 7-X has recently proven to be at par with tokamak plasma performance for long discharges. This talk provides a brief review of the principles of nuclear fusion and discusses the key physics subjects of optimized stellarators. We summarize the most important findings of the previous performance operation campaigns and put them into the international context of fusion research. An outlook is given towards fusion power as a building block of future energy supply of the world. [1] National Geographic, November Issue 2025 https://www.nationalgeographic.com/science/article/stars-nuclear-fusion-energy [1] T. Klinger et al., Nuclear Fusion 2019, 59(11) 112004 [2] A. Dinklage et al., Nature Physics 2018, 14(8) 855

Teilchenkolloquium

CP asymmetry in D0?K0SK0S

Dr. Giulia Tuci, Physikalisches Institut, Uni Heidelberg

Astronomisches Kolloquium

Dienstag, 18. November 2025 16:30 Uhr  From Data to Laws: Symbolic Regression and Differentiable Analytic Networks for (Astro)physics

Rodrigo Ibata, Observatoire Astronomique de Strasbourg Over the next few years, Rubin/LSST, Euclid, Roman, SKA, and other instruments will produce petascale, information?rich datasets that trace stars, galaxies, and large-scale structure with unprecedented fidelity. Hidden in these data may be regularities that point to new and unexpected physical relationships. Can we build modelling frameworks that can discover such relationships accurately, efficiently, and in forms we can interpret? I will present two complementary directions we are developing to address this question. The first, PhySO, is a physics-aware symbolic regression engine which proposes compact mathematical equations using deep reinforcement learning with a dimensional-analysis grammar and imposable constraints. The second, NestyNet, assembles networks with analytic derivatives and trains them using second-order methods, yielding fast, high accuracy fits to datasets, solvers for ODEs/PDEs, action-angle transformations, Gaussian-mixture inference, and dynamical modelling, with exact gradients and Hessians throughout. I will demonstrate how symbolic search coupled with accurate derivatives and with PDE constraints can rediscover analytic solutions from textbook physics. This approach is a practical route toward explainable, robust models for the forthcoming data deluge-aimed less at "automating Kepler" than at accelerating analysis while keeping physical insight. To arrange a visit with the speaker during the visit, please contact their host: Morgan Fouesnau

Zentrum für Quantendynamik Kolloquium

Mittwoch, 26. November 2025 16:30 Uhr  tba

Prof. Guido Pupillo, Institut de science et d'ingénierie supramoléculaires, Unversité de Strasbourg