Drift instabilities, anomalous transport, and heating in low-temperature plasmas

Abstract

Plasma is an ideal gas of charged particles (ions and electrons) in addition to neutral particles. The presence of charged particles results in the generation of electric and magnetic fields that serve as the primary mechanism of the interaction and coupling of particles. As a result, various nonlinear collective phenomena occur in the plasma, the understanding of many of which remains elusive today. On the other hand, plasmas have many applications in different branches of science and technology. Different kinds of plasmas are studied in the atmospheric and space sciences. In the semiconductor industry, the fabrication of electronic chips relies heavily on plasma etching. Plasma is used in modern electrical thrusters for producing the driving force of satellites and spacecrafts. It is also used in future fusion reactors for producing abundant clean energy. Therefore, understanding the complicated phenomena in plasma is important for predicting and controlling its behaviours in various conditions. In this regard, nonlinear phenomena, such as turbulence, are formidable barriers to understanding plasma behaviours. These phenomena are described by nonlinear differential equations that can be barely understood by analytical means and are usually investigated by numerical simulations. Because of this, it is also important to understand the effect of numerical artifacts on simulations. In this thesis, we investigate the nonlinear characteristics of drift instabilities and the role of numerical methods in our understanding of these instabilities. The drift instabilities are driven by excess free energy that exists due to the average (drift) velocities of electron and ion components in plasmas. As a result of these instabilities, the amplitude of fluctuations grows while the drift energy converts into electrostatic energy. This growth continues until the nonlinear effects, such as turbulence, trapping, and wave-wave interactions, become active. As a result of these nonlinear effects, the growth of the fluctuations saturates. In this thesis, our focus will be on two particular types of drift instabilities, namely the Buneman instability and electron-cyclotron drift instability (ECDI). The Buneman instability is driven when a beam of electrons is injected into the stationary ions, while both electrons and ions are unmagnetized. In the ECDI, however, the electrons are magnetized and are also influenced by an external electric field, perpendicular to the magnetic field. This configuration of fields leads to the E × B drift of the electrons that drives the ECDI. Many kinetic simulations are performed, and several nonlinear phenomena such as trapping, heating, anomalous transport, backward waves, and transition of magnetized plasmas to the unmagnetized regime are studied with regard to both instabilities. For the study of the nonlinear effects of drift instabilities, a grid-based Vlasov code is developed and used. The numerical method used in this code is the “semi-Lagrangian” method, which is among the most popular methods for continuum simulations of plasma. In the study of the drift instabilities, we compare the results of the semi-Lagrangian Vlasov simulations with the more traditional particle-in-cell (PIC) method. The results of these benchmarking studies reveal several similarities and discrepancies between Vlasov and particle-in-cell simulations, showing how the numerical methods can interfere with the physics of the problems.