Collisions in strongly magnetized pair plasmas & optimization and testing of non-planar stellarator coils using high-temperature superconductors
Year: 2026
Type: dissertation
Abstract: Pair plasmas made from electrons and their antiparticles, positrons, are expected to occur in extreme astrophysical environments and are predicted to enjoy remarkable stability and the absence of turbulence. However, they have never been confined and studied for longer periods in a laboratory setup on Earth. In this thesis, both theoretical and experimental work are presented to facilitate the confinement of an electron-positron plasma in a tabletop, optimized stellarator called EPOS (R=0.19\,m, a = 0.051\,m). To create a pair plasma, up to $10^{10}$ positrons from the NEPOMUC positron source at the FRM II research fission reactor can be accumulated in buffer traps, and cooled to a temperature of $\leq 5\,$eV before being injected into EPOS. Combined with electrons, they are expected to form a plasma with a density of $10^{12}\,$m$^{-3}$. If the magnetic field is high enough ($2\,$T), cyclotron cooling will cool the perpendicular energy of the particles away. Plasmas in EPOS are expected to be in the strongly magnetized regime. The Larmor radius is much larger than the Debye length. For most plasmas in fusion research and industry, the opposite is true. In a strongly magnetized plasma, the magnetic field cannot be neglected during particle collisions, necessitating a numerical treatment. To assess collisions in EPOS, we perform simulations of binary particle collisions at expected plasma parameters. From these, we calculate the collisional step size and classical and neoclassical diffusion coefficients. The results indicate collisions do not significantly affect plasma transport or confinement in EPOS. At low temperatures ($\leq 10$ meV), temporarily correlated electron-positron pairs (magnetobound positronium) form in low-impact-parameter collisions. These pairs can drift across magnetic field lines for many Larmor radii, thereby increasing radiative recombination of electrons and positrons into positronium. The high magnetic field required for EPOS motivates the use of ReBCO (Rare-Earth Barium Copper Oxide) high-temperature superconductors, which exhibit high current density at EPOS magnetic fields and can be cooled by a cryocooler instead of liquid helium. However, ReBCO tape is sensitive to torsion and bending strains, which may prevent it from being wound into the intricate, non-planar shape of stellarator coils. Additionally, ReBCO shows strong critical current anisotropy with respect to the angle between its c-axis and the magnetic field. In this thesis, we incorporated penalty functions into the stellarator optimization framework SIMSOPT to optimize both the shapes and winding pack orientations of stellarator coils. This lowers strain and increases the critical current. These improvements avoid damage to the superconducting tape and increase coil performance. The strain penalty enables the optimization of a coil set at the small size of EPOS compatible with the use of ReBCO. Both optimization targets are also applied to other stellarators, from university-scale to reactor-scale devices. To guide EPOS magnet construction, prototype coils were built and tested at 77 K in liquid nitrogen and at temperatures of 20 K or below using a cryocooler. Pathfinder coils provided experience with ReBCO. A non-planar coil at the EPOS scale showed no tape defects, validating strain optimization. A planar, high-field coil exceeded 3 T field strength at the superconductor, reaching EPOS winding pack conditions. Lessons from earlier tests informed the design of the EPOS model coil. A full-size, full-current, non-planar coil was tested in liquid nitrogen and now awaits testing at 20 K.
DOI: 10.5282/edoc.37310
Author: Huslage, Paul
Institution: Ludwig-Maximilians-Universität München
Language: English
Subfield: Nuclear and High Energy Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7214915183