Search

The plasma edge region has a significant influence on the magnetic confinement in fusion plasmas, which in turn determines the efficiency of a possible future fusion reactor. Thus, the processes taking place in this region have to be understood in detail to be able to accurately control the edge region. A numerical way to gain deeper insights into the complex processes within the plasma sheath region is the ab intio kinetic particle simulation. The kinetic description is capable of resolving even the smallest scales of interest in time and space. The Particle-in-Cell scheme is a grid based representative of this class of simulations. It has been used for simulations of the plasma sheath for a considerable time. Within this work, another concept of kinetic particle simulations is used – the Barnes-Hut Tree Code. This method has been successfully used for simulations of laser plasmas as well as for simulations of galaxy formation before. Due to intrinsic properties of this method, it should be well suited for treating typical plasmas in the plasma wall interaction region. However, contradictions to analytical as well as numerical results were found in previous attempts to simulate the edge region with a Tree Code. These contradictions had not been resolved so far. In the present work the massively parallel Barnes-Hut Tree Code PEPC is used to develop and evaluate a numerical model of the plasma wall interaction region. The discrepancies have been understood and completely dispelled. An artificially increased collisionality within the simulation plasma and numerical heating of the electrons were identified as the cause aforementioned contradictions. For the first time, these effects have been measured and parametrized by plasma density, plasma temperature and several numerical input parameters in a wide parameter range. Using the obtained results, the electron and ion kinetics within the plasma sheath region have been self-consistently modeled from first principles as a proof of concept.Given this proof of concept, we developed a guideline that helps to evaluate, if a given plasma edge problem can be tackled with the grid free particle simulation. This guideline is based upon two criteria. First, the aforementioned parametrization of the collision and heating times in a simulation plasma as well as additional stability criteria are used to exclude numerical parameters that lead to incorrect results. Second, considerations regarding the runtime of the massively parallel simulations are used to define a maximum system size and simulation length. Finally, usage of the developed guideline is demonstrated by means of several relevant examples.

Intrastromal laser surgery with picosecond pulses aims to achieve refractive changes of the cornea without damaging the epithelium, Bowman's membrane, or the endothelium. For that, a tissue layer with well-defined thickness has to be evaporated by creating laser plasmas within the corneal stroma. We investigated the plasma formation and the plasma-induced shock wave emission and bubble generation (cavitation) in the cornea, as well as the tissue effects and the range for endothelial damage. The laser light source used was an Nd:YAG laser emitting pulses with a duration of 30 ps at a repetition rate of 10 Hz. Intrastromal plasma formation and cavitation were investigated in sheep eyes in vitro by means of time-resolved macro-photography with 20 ns exposure time. Photographs were taken at various delay times (3 microseconds-2 min) after the release of the Nd:YAG laser pulse. The morphology of the laser effects and the incidence of endothelial damage was investigated by light-microscopic inspection of histological cross sections of the irradiated corneas. The minimal plasma size at energies close to the breakdown threshold was about 40 microns. Using a laser effects could be created without causing microscopically detectable damage to the epithelium, endothelium, or Bowman's membrane. To avoid damage, the distance between endothelium and laser focus had to be larger than 150 microns. Shock wave-induced tissue damage was not observed, although the maximum shock wave pressure was up to 13 kbar. The laser-generated intrastromal cavities are at least 10 times larger than the plasma volume.(ABSTRACT TRUNCATED AT 250 WORDS)