Development of mathematical model representing dynamics of plasma generated from a plasma focus device

Abstract

Plasma Focus (PF) device is known to be used as source of radiations and particles. It also has been used for materials surface modification. This research aims to investigate the dynamics of plasma generated by a standard UNU/ICTP PF device. More detailed microscopic modeling of the dynamics is attempted, which is different and can be complimentary to the widely used Lee model code. Lee model is considered to be a macroscopic model as it assumed that plasma behave as a thin solid cylinder that moves by Lorentz force. The Lorentz force is generated by the plasma current and the induced magnetic field from the discharge current where the circuit equation is used to represent the discharge characteristic. In this research, a software package is used to simulate the dynamics of plasma microscopically using finite element method. It considers geometry of PF, material properties, electromagnetic theory, plasma theory and input current characteristic for the finite element calculation. Both simulation results are compared with experimental results, where UNU/ICTP PF is operated with variable pressure of argon gas at 1.0, 1.5 and 2.0 mbar. Current factor and mass swept factor used by Lee model are found to be between 0.36-0.67 and 0.024-0.0365 respectively. The finite element simulation gives average current factor of 0.48-0.55. It is found to be varying with time. The finite element simulation also has shown that the density and the temperature of argon gas changes as plasma is moving along the electrodes. However, it has shown that not all argon gas is swept by the plasma which correspond to the assumption of mass swept factor applied in Lee model. In addition, the finite element simulation uses plasma processes to represent the behavior of the plasma generated by PF device microscopically, which allows graphical generation of plasma starting from breakdown phase to axial phase where Lee model cannot show. It has been demonstrated that a microscopic model can generate results and characteristics of plasma that are close to actual experiment. They also agree with the results from the macroscopic model based on Lee model code.