Particle Methods for Modeling Magnetospheric Diagnostics and Low-Temperature Plasma Physics

Abstract

Single-particle-motion codes trace the evolution of charged particles within applied electromagneticfields. These numerical tools are widely applicable to study the evolution of six-dimensional phase-space distribution functions, with three space and three velocity dimensions. Coupled with grid based methods for solving the self-consistent electrostatic fields these techniques can be extended to study complex kinetic plasma phenomena through the particle-in-cell method. This thesis is broadly split into two components. The first sections of Chapter 1 details the applied mathematical aspects of single-particlemotion codes as well as the particle-in-cell method, and Chapters 2 through 4 apply these techniques to problems in magnetospheric physics and low-temperature plasmas. The second half of Chapter 1 and Chapter 5 discuss the development of a new particle-in-cell code to enable whole-device modeling of low-temperature plasma devices. In Chapter 2 a single-particle-motion code is applied to study the evolution of a proposeddiagnostic for probing the Earth’s magnetosphere. Tracing magnetic field-lines of the Earth’s magnetic field using beams of relativistic electrons will open up new insights into space weather and magnetospheric physics. Analytic models and the single-particlemotion code are used to explore the dynamics of an electron beam emitted from an orbiting satellite and propagating until impact with the Earth. It is observed that the impact location of the beam on the upper atmosphere is strongly influenced by magnetospheric conditions, shifting up to several-degrees in latitude between different phases of a simulated solar storm. The beam density cross-section evolves due to cyclotron motion of the beam centroid and oscillations of the beam envelope. The impact density profile is ring shaped, with major radius ∼ 22 meters, given by the final cyclotron radius of the beam centroid, and ring thickness ∼ 2 meters given by the final beam envelope. Motion of the satellite may also act to spread the beam, however it will remain sufficiently focused for detection by ground-based optical and radio detectors. An array of such ground stations will be able to detect shifts in impact location of the beam, and thereby infer information regarding magnetospheric conditions. Chapter 3 applies a kinetic particle-in-cell code to investigate the formation of rotatingspokes within a Penning discharge. This instability is commonly observed within E × B discharges, such as Hall thrusters or magnetrons, where it can lead to non-uniform and inefficient operation. Electron cross-field transport within the Penning discharge is highly anomalous and correlates strongly with the spoke phase. Similarity between collisional and collisionless simulations demonstrates that ionization is not necessary for spoke formation. Parameter scans with discharge current Id, applied magnetic field strength B and ion mass mi show that spoke frequency scales with sqrt(eErLn/mi), where Er is the radial electric field, Ln is the gradient length scale and e is the fundamental charge. This scaling suggests that the spoke may develop as a non-linear phase of the collisionless Simon-Hoh instability. Incoherent Thomson scattering is a non-intrusive laser diagnostic commonly used formeasuring local plasma density. In Chapter 4, the same particle-in-cell code is applied to demonstrate that within low-density, low-temperature plasmas and for sufficient laser intensity, the laser may perturb the local electron density via the ponderomotive force, causing the diagnostic to become intrusive and leading to erroneous results. These results support predictions offered in a recent theoretical paper. The second part of Chapter 1 introduces some of the computational challenges associatedwith obtaining performance with a particle-in-cell code on modern supercomputers. These include exposing on-node parallelism via memory sharing paradigms as well as acceleration with GPUs. The mixed particle-grid nature of the algorithm leads to a range of choices for parallelisation across multiple nodes, with the advantages and disadvantages of each method discussed. Chapter 5 justifies the development of a new particle-in-cell code (LTP-PIC), with a focus on capabilities relevant to whole device modeling of lowtemperature plasma devices. This new code is applied to directly study the effects discussed in Chapter 1, and how the multi-level hierarchy of memory and parallelism on modern supercomputers can be leveraged to achieve performance. The code is also benchmarked against six other international codes, and applied to model a simple three-dimensional extension of that in Chapter 3. Chapter 5 may leave the reader wondering what happens next? Indeed LTP-PIC is awork in progress, and while it is well tuned to take advantage of modern supercomputers, it requires additional features to be broadly applicable to a wide range of low-temperature plasma devices. It is the author’s intention to continue pursuing development of this software as a platform for innovation, industry collaboration and as an open-source tool for the low-temperature plasma community.