Multiscale Modeling of Hall Thrusters Open Access
Downloadable ContentDownload PDF
Hall thrusters are efficient space propulsion devices that produce thrust by accelerating quasineutral plasma. The thruster consists of a channel with one end open to the ambient environment. The neutral propellant is injected at the closed end and is ionized by electrons produced by an externally mounted cathode. A radial magnetic field is applied across the channel to restrict the electron flow and increase the ionization efficiency. The magnetic field also become lines of constant potential and induce an electric field that accelerates the ions out of the device.This thesis introduces a novel approach for modeling these devices. Hall thrusters have been operated in space for over 40 years, yet no code yet exists that can self-consistently predict their operation and lifetime. This shortcoming is to a large extent driven by the presence of significantly different spatial scales that dominate the internal plasma dynamics. One such a scale is the microscopic scale of an electron orbiting about a magnetic field line. This scale determines the rate with which electrons diffuse across the field line which becomes an important parameter in modeling the internal thruster discharge. It is not numerically feasible to include electron motion when dimensions of a thruster are considered. As such, Hall thruster codes typically utilize a hybrid approach in which electrons are modeled as a fluid with electron transport given by an analytical expression. These expressions do not correctly capture the experimentally-observed behavior. In this work, a kinetic code was developed to determine transport at the desired spatial locations self-consistently by considering the rate with which electrons diffuse across a magnetic field line. This kinetically-determined transport can then be utilized in the discharge code instead of the analytical model. This code is utilized in this work to study fundamentals of electron transport. Among other findings, simulation results indicate that the linear combination of classical and wall induced transport is not valid due to synergistic effects.Thruster codes also do not accurately capture the near-wall region known as the sheath which drives ion flux to the walls. In addition, often the plume produced by these devices needs to to be analyzed in order to quantify any possible spacecraft contamination events. On the scale of the spacecraft, the thruster internal details are no longer resolved and the thruster acts as a source of ions and neutrals into the plume analysis. These two topics describe the other two spatial scales studied in this work. The near-wall region is analyzed with a particle code which resolves the near-wall non-neutral region and also estimates wall erosion rates based on ion impacts. This code also takes into account magnetic field line inclination. At highly-inclined magnetic field line angles, the code predicts that ions become repelled from the wall, resulting in a sheath collapse. Finally, a novel contribution to the plume modeling effort is a time-dependent source that injects particles based on a time-resolved kinetic sampling of particles leaving the thruster simulation. Hall thrusters are inherently unsteady devices, however, typical plume codes assume steady operation. Preliminary results indicate that the unsteady operation of these devices is demonstrated in a corresponding oscillating current of backflowing ions.