Design and Assessment of a Standalone Continuous Toroidal Electromagnet for Novel Stellarator Concepts

Abstract

Fusion has the potential to be a paradigm-shifting energy generation source with the capability to accelerate both the transition to renewable energy and meet growing global demand. Recent advances in the computational optimization of one fusion reactor design, the stellarator, have produced the need for various physical experiments to study their plasmas. However, building a new stellarator is a complex, expensive, and time-consuming process, largely due to the manufacturing challenges of their nonplanar magnetic field coils. As a result, it is cost prohibitive to build multiple stellarator experiments. By reevaluating the coil design, a single experimental machine could be built with the ability to generate the magnetic field of many unique stellarator designs, lowering the investment required to study a range of optimized plasmas. This thesis presents a method of manufacturing a continuous 2D "surface coil" for a stellarator to replace the typical set of discrete coils. To this end, a twelfth torus stellarator coil was built out of Galinstan using a plastic mold. The magnetic flux density at coil currents of 300 and 900 A was measured at 128 distinct points by a set of Hall effect sensors on a custom printed circuit board. Compared with simulation in COMSOL Multiphysics, the measurements in the axial direction of the coil had maximum, median, and mean errors in the Y direction of 47.72\%, 12.79\%, and 9.51\% for 300 A and 18.84\%, 3.61\% and 4.76\% for 900 A. For the other directions, these errors were one to two orders of magnitude higher, likely attributable to the influence of unshielded wires external to the coil. Although the error in the magnetic flux density for a real-world fusion device should be less than one percent, this thesis made a significant step towards demonstrating the feasibility of a 2D "surface coil."