Democritus: a simulation tool for plasma-material interaction

Summary form only given. DEMOCRITUS is a particle-based software infrastructure for plasma-material interaction simulation. The fundamental approach is the use of particle in cell (PIC) methods to simulate the plasma species, the material and their interaction. DEMOCRITUS is based on three ingredients: 1) particles to treat all species in any phase they are: solid, liquid, gas or plasma; 2) an adaptive grid to describe the interaction fields; 3) the immersed boundary method to describe the interface among different materials. The particles are standard computational particle with a finite size defined by a shape function. Linear shape functions are used. Each material is represented by different particles, label in DEMOCRITUS by a integer color variable. The particles have many properties, besides the usual position, velocity and charge, to describe the material properties. The particle description is adaptive, allowing the increase or decrease of number of particles locally to increase accuracy. The grid is 2D but can be either in cylindrical coordinates (r,z) or Cartesian (x,y). An adaptive grid is used based on a new implementation of the Brackbill-Saltzmann variational grid adaptation method that minimizes an a-posteriori definition of the error in the simulation. The immersed boundary method is a widely used approach to translate boundary or interface conditions among material into a formalism that allows to use a continuum model without any discontinuity in the formulation of the equations. We use a first order formulation of the method, and rely on the adaptive grid to control the truncation error in describing any localized feature at material boundaries. We present two typical applications of DEMOCRITUS: the charging of a small object (dust particle) immersed in a plasma and the focusing of a spherical electron population in the Los Alamos inertial electrostatic confinement (IEC) fusion device