Joseph Wang

Forward Cascade of Whistler Turbulence: Three-dimensional Particle-in-cell Simulations

This manuscript describes the first ensemble of three-dimensional (3D) particle-in-cell (PIC) plasma simulations of whistler turbulence. The computational model represents a collisionless, homogeneous, magnetized plasma on which an initial spectrum of relatively long wavelength whistler fluctuations is imposed. The simulations represent a range of initial fluctuation amplitudes and follow the temporal evolution of the system as it decays into a broadband, anisotropic, turbulent spectrum at shorter wavelengths via a forward cascade. The resulting 3D turbulence is similar in many ways to whistler turbulence from previous two-dimensional (2D) PIC simulations, although the anisotropies in 3D are stronger than in comparable 2D runs. The most important difference is that reduced magnetic fluctuation spectra from the 3D simulations show a clear break in the perpendicular wavenumber (k ⊥) spectra. Spectra at small k ⊥ are relatively steep, but spectra at larger k ⊥ are even steeper, similar in character to magnetic spectra at electron scales recently measured in the solar wind.

Three dimensional character of whistler turbulence

It is shown that the dominant nonlinear effect makes the evolution of whistler turbulence essentially three dimensional in character. Induced nonlinear scattering due to slow density perturbation resulting from ponderomotive force triggers energy flux toward lower frequency. Anisotropic wave vector spectrum is generated by large angle scatterings from thermal plasma particles, in which the wave propagation angle is substantially altered but the frequency spectrum changes a little. As a consequence, the wave vector spectrum does not indicate the trajectory of the energy flux. There can be conversion of quasielectrostatic waves into electromagnetic waves with large group velocity, enabling convection of energy away from the region. We use a two-dimensional electromagnetic particle-in-cell model with the ambient magnetic field out of the simulation plane to generate the essential three-dimensional nonlinear effects.

Modeling Electrostatic Levitation of Dust Particles on Lunar Surface

This paper presents a simulation model on electrostatic levitation of lunar dust particles in the lunar terminator region. Full-particle particle-in-cell simulations are carried out using real ion to electron mass ratio to obtain plasma sheath, surface charging, and the transition point of surface electric field. Test particle simulations are carried out to simulate the levitation of dust particles from lunar surface. Results show that the dust levitation condition in the terminator region is sensitively influenced by the ambient plasma condition and surface charging, and the levitation altitude varies significantly even for small changes of the sun elevation angle.

Development of the DRACO Code for Modeling Electric Propulsion Plume Interactions

COLISEUM is a plasma simulation package for modeling electric plume interactions currently being developed by Air Force Research Lab, MIT, and Virginia Tech. One of the major components of COLISEUM is the DRACO code. DRACO, designed to perform firstprinciple based, high fidelity simulations of plume spacecraft interactions, include several simulation modules. As the purpose of DRACO is to simulate real engineering problems, DRACO reads in spacecraft configurations defined by commercial CAD tools. A unique feature of DRACO is that it incorporates the recently developed immersed finite element particle-in-cell (IFE-PIC) algorithm. This method allows one to use a Cartesian mesh to handle realistic spacecraft geometry without scarifying the accuracy in electric field solutions. The computational speed of an IFE-PIC simulation is about the same as that of standard PIC simulation. DRACO is cross platform and runs on Windows, Linux, and Unix. This paper presents an overview of the DRACO code and presents simulation results for simulation of CEX backflow in a vacuum tank as well as plume/spacecraft interaction in the presence of a charged plume shield. The Air Force Research Laboratory (AFRL) is sponsoring the development of a flexible, user-friendly, plasma computational package called COLISEUM. The core library of COLISEUM provides the rudimentary input/output support to several plasma simulation packages. The complexity of the simulation packages ranges from a simple ray-tracing algorith, through a prescribed plume model to several ES-PIC simulation modules. The DRACO module, being developed at Virginia Tech, is described in this paper. Additional information about COLISEUM and its simulation packages is available in Ref. 1-2. DRACO is a multi-purpose electro-static, particle-in-cell (ES-PIC) plasma simulation package. As shown in Fig. I, DRACO allows the user to choose from several Poisson solvers. The quasi-neutral solver obtains the potential by assuming the Boltzmann distribution for the electron density and a constant electron temperature. The quasi-neutral solver is intended for quick calculations and cannot be used to resolve the plasma sheath. The DADI solver uses a standard finite-dierence formulation to solve the electric field. It is designed for problems with relatively simple geometries. The Immersed Finite Element (IFE) is DRACO’s most sophisticated solver. IFE is based on a finite element formulation, and is designed to perform simulations accurately for problems involving complex geometric and material boundary conditions. Instead of using a complex body-fitted mesh, the IFE method uses a structured mesh without consideration of the object surface location. Thus the standard, Cartesian mesh based method for particle-mesh interpolations and pushing particles can be used even in simulations involving complex geometric boundaries. This allows DRACO to retain the computation speed of a standard PIC code. Many of DRACO s subroutines are based on 3-D plasma simulation codes previously developed by J. Wang to simulate ion thruster plume interactions 4 and ion optics plasma flow. 5 The legacy code produced results that were in excellent agreement with data from Deep Space 1 in-flight measurements and the long duration test of the NSTAR thruster.

A Hybrid Grid Immersed Finite Element Particle-in-Cell Algorithm for Modeling Spacecraft–Plasma Interactions

This paper presents a new algorithm for modeling spacecraft-plasma interactions: the hybrid grid immersed finite element particle-in-cell (HG-IFE-PIC) algorithm. This algorithm is extended from IFE-PIC and is designed to efficiently and accurately simulate the interactions between nonuniform plasmas and complex objects. In HG-IFE-PIC, the meshes used by the IFE field solver and PIC are decoupled, and the IFE solver uses a Cartesian-tetrahedral mesh that is stretched according to local potential gradient and plasma density. It is shown that the application of HG-IFE-PIC with a stretched IFE mesh can achieve approximately the same accuracy as IFE-PIC while significantly reducing the memory requirement and computation time required for large-scale PIC simulations of spacecraft-plasma interactions

Simulations of Ion Thruster Plume–Spacecraft Interactions on Parallel Supercomputer

A parallel three-dimensional electrostatic Particle-In-Cell (PIC) code is developed for large-scale simulations of ion thruster plume-spacecraft interactions on parallel supercomputers. This code is based on a newly developed immersed finite-element (IFE) PIC. The IFE-PIC is designed to handle complex boundary conditions accurately while maintaining the computational speed of the standard PIC code. Domain decomposition is used in both field solve and particle push to divide the computation among processors. A high-resolution simulation of multiple ion thruster plume interactions for a realistic spacecraft using a domain enclosing the entire solar array panel is carried out to demonstrate the capability of the code. A standard finite-difference PIC (FD-PIC) code is also parallelized for comparison. Both the IFE-PIC and FD-PIC run with similar overall speed, and the IFE-PIC achieves a high parallel efficiency of ges 90%

Electron–Ion Coupling in Mesothermal Plasma Beam Emission: Full Particle PIC Simulations

Full particle particle-in-cell simulations are performed to study the collisionless electron-ion coupling during ion beam emission and neutralization. Simulations show that ion beam neutralization and propagation are two closely coupled processes. Electron-ion coupling is achieved through interactions between the trapped electrons and a potential well established by the propagation of the ion beam front along the beam direction and not through plasma instabilities as previous studies suggested. In the transverse direction, the beam emission generates an expansion fan similar to that of the expansion of a mesothermal plasma into vacuum. The expansion process determines the beam potential with respect to the ambient. This suggests that a plasma beam in a vacuum chamber and a plasma beam in space may have similar charge density profiles but different beam potentials with respect to the ambient due to the limit imposed by the boundary on plasma expansion.

Whistler anisotropy instability: Spectral transfer in a three‐dimensional particle‐in‐cell simulation

A three-dimensional (3-D) particle-in-cell (PIC) simulation of the whistler anisotropy instability is carried out for a collisionless, homogeneous, magnetized plasma with βe=0.10. This is the first 3-D PIC simulation of the evolution of enchanced fluctuations from this growing mode driven by an anisotropic electron velocity distribution with T⟂e/T∥e>1 where ⟂ and ∥ represent directions perpendicular and parallel to the background magnetic field Bo, respectively. The early-time magnetic fluctuation spectrum grows with properties reflecting the predictions of linear theory with narrowband maxima at kc/ωe≃ 1 and k×Bo=0, and a wave vector anisotropy in the sense of k⟂ >k∥. The inverse transfer is consistent with a prediction of nonlinear three-wave coupling theory.

A 3D immersed finite element method with non-homogeneous interface flux jump for applications in particle-in-cell simulations of plasma-lunar surface interactions

Motivated by the need to handle complex boundary conditions efficiently and accurately in particle-in-cell (PIC) simulations, this paper presents a three-dimensional (3D) linear immersed finite element (IFE) method with non-homogeneous flux jump conditions for solving electrostatic field involving complex boundary conditions using structured meshes independent of the interface. This method treats an object boundary as part of the simulation domain and solves the electric field at the boundary as an interface problem. In order to resolve charging on a dielectric surface, a new 3D linear IFE basis function is designed for each interface element to capture the electric field jump on the interface. Numerical experiments are provided to demonstrate the optimal convergence rates in L 2 and H 1 norms of the IFE solution. This new IFE method is integrated into a PIC method for simulations involving charging of a complex dielectric surface in a plasma. A numerical study of plasma-surface interactions at the lunar terminator is presented to demonstrate the applicability of the new method.

Virtual Anode in Ion Beam Emissions in Space: Numerical Simulations

The plasma interactions associated with the emission of an unneutralized ion beam in space are discussed. Three-dimensional particle simulations, which follow the beam ions as well as the ambient electrons and ions in the self-consistent electric e eld, are performed for various emission conditions. Classical theories and laboratory experiments for one-dimensional beam e ow in diodes have shown that a potential hump will form when the beam current density exceeds a critical value. When the potential hump becomes so high that the beam kinetic energy is near zero there, it becomes a virtual electrode. We show that similar characteristics are also exhibited in high-density ion beams emitted from spacecraft to space plasma, although three-dimensional effects and the ambient plasma make the interaction more complex. Once a virtual anode has formed, it will partly block the beam transmission and discharge the spacecraft potential.