Particle-In-Cell Modeling of Plasma-Based Accelerators in Two and Three Dimensions

In this dissertation, a fully object-oriented, fully relativistic, multi-dimensional Particle-In-Cell code was developed and applied to answer key questions in plasma-based accelerator research. The simulations increase the understanding of the processes in laser plasma and beam-plasma interaction, allow for comparison with experiments, and motivate the development of theoretical models. The simulations support the idea that the injection of electrons in a plasma wave by using a transversely propagating laser pulse is possible. The beam parameters of the injected electrons found in the simulations compare reasonably with beams produced by conventional methods and therefore laser injection is an interesting concept for future plasma-based accelerators. Simulations of the optical guiding of a laser wakefield driver in a parabolic plasma channel support the idea that electrons can be accelerated over distances much longer than the Rayleigh length in a channel. Simulations of plasma wakefield acceleration in the nonlinear blowout regime give a detailed picture of of the highly nonlinear processes involved. Using OSIRIS, we have also been able to perform full scale simulations of the E-157 experiment at the Stanford Linear Accelerator Center. These simulations have aided the experimentalists and they have assisted in the development of a theoretical model that is able to reproduce some important aspects of the full PIC simulations. Update (2015): This dissertation was originally written in 2000. I am making it now available on arXiv with the hope that some its content might proof useful to the users of the OSIRIS code which has continued to be utilized by a number of research groups since it was originally written as part of the research presented in this dissertation.

💡 Research Summary

This dissertation, completed in 2000 by Roy Gerrit Hemker at UCLA, presents the development and application of “OSIRIS,” a fully relativistic, object-oriented, multi-dimensional Particle-In-Cell (PIC) code, for modeling plasma-based accelerators. The work addresses key questions in the field by providing high-fidelity simulations of complex laser-plasma and beam-plasma interactions.

The core achievement is the creation of the OSIRIS code itself. It was designed from the ground up using object-oriented principles, enabling a unified framework for simulations in 2D Cartesian, 3D Cartesian, and 2D cylindrically-symmetric geometries. This design emphasized flexibility, maintainability, and portability to parallel computing architectures, featuring advanced capabilities like dynamic simulation spaces (e.g., moving windows) for modeling phenomena that propagate over long distances.

The dissertation applies OSIRIS to several pivotal research areas in plasma acceleration. First, it investigates a “cathodeless injector” concept, where a transversely propagating laser pulse intersects a primary laser wakefield driver to inject electrons into the accelerating phase of the plasma wave. Both 2D and 3D simulations demonstrated the feasibility of this method, showing that it could produce electron beams with qualities comparable to conventional injectors.

Second, the research explores the “long wavelength hosing” instability of a laser pulse propagating in a plasma, providing simulation evidence that complements theoretical models. Third, simulations of Laser Wakefield Acceleration (LWFA) in a parabolic plasma channel show that the laser driver can be optically guided over distances far exceeding its vacuum Rayleigh length, enabling sustained electron acceleration.

A major application involved full-scale simulations of the “E-157” Plasma Wakefield Acceleration (PWFA) experiment at the Stanford Linear Accelerator Center (SLAC). OSIRIS was used to model the highly nonlinear “blowout” regime where a dense electron driver beam expels all plasma electrons from its path. These simulations offered a detailed microscopic picture of the beam and plasma dynamics, directly aiding experimentalists in interpreting their results and facilitating the development of a simplified analytical model that captured key aspects of the full PIC simulations.

In summary, this dissertation made a dual contribution: it delivered a sophisticated, state-of-the-art PIC code (OSIRIS) that became a foundational tool for the plasma accelerator community, and it employed this tool to gain fundamental insights into injection schemes, laser propagation instabilities, and the nonlinear dynamics of both laser-driven and beam-driven plasma wakes. The 2015 update notes the code’s continued use and relevance, underscoring the lasting impact of this work.