Experimental Assessment of Plasma Transport in Magnetic Multicusp Ion Sources.

Abstract

The objective of this dissertation is to experimentally analyze plasma transport to the anode in multicusp discharge chambers and evaluate its impact on discharge performance. The physics of plasma transport from the bulk plasma through the magnetic cusp to the anode remains poorly understood. A proper accounting of plasma losses to the anode is critical to accurate modeling of multicusp device performance. In such models, plasma losses are balanced with current draw to determine plasma parameters. Previous studies have put much effort into quantifying the leak width, which is the width of the line of collection that runs along the interior of each cusp. The conventional assessment of plasma losses in multicusp devices is to multiply the area of collection by the incident current density of each particle species. This assessment fails to take into account magnetic field effects, collisional effects, and electrostatic effects that impact current flow in the cusp region. A more accurate and thorough accounting of plasma losses in multicusp devices is needed to improve predictive models.

Ring current measurements coupled with spatially resolved plasma parameters throughout a 16-cm multicusp discharge chamber enabled an assessment of plasma losses to each ring in terms of an ”effective loss area” which, multiplied by electron current density incident on the bulk/cusp boundary, gave the correct current to each ring. The effective loss area was compared to the physical loss area at the rings by a correction factor. Plasma transport in the cusps was studied by mapping electron density in the region above a magnetized anode using laser-collisional induced fluorescence. The leak widths were found to scale with the hybrid gyroradius. As predicted by theory, the constant of proportionality was dependent on chamber pressure and, by extension, the electron and ion mean free paths. These findings were used to improve a predictive 0-D particle and energy balance model.