A high power pulsed plasma system for material testing under simultaneous continuous and transient loads

Abstract

The research described in this thesis is done in the frame of the development of nuclear fusion as a source of clean, safe and virtually inexhaustible power. Fusion energy is being developed in an international program, which is presently culminating in the construction of the test reactor ITER. ITER is designed to produce ten times more power than is needed to run it, at a level of 500 MW (comparable to a small power plant). In ITER a plasma of hydrogen isotopes is confined by strong magnetic fields and heated to 250_106 _C. The basics of the magnetic confinement and heating are well understood, but there are other areas where ITER enters uncharted waters. One of these is the interaction that takes place where the plasma meets the wall of the reactor: plasma surface interaction(PSI), the topic of this thesis. The high power density in the reactor leads to potential problems with melting, evaporation and erosion of the wall material, as well as the retention of hydrogen either in the wall itself or in redeposits of eroded material. In ITER, the fluences of particles and energy to the wall are two orders of magnitude larger than those in earlier experiments, and if the PSI and materials issues are not dealt with properly, the life time of the strike zones i.e. the strips of material that take most of the power, could be days rather than years. The problem of the power exhaust is twofold. To start, the time averaged heat flux density to the strike zone is already extreme, at 10 MW_m2. Yet, materials and cooling techniques have been developed that can handle that power load if it comes as a steady ux. However, the situation is aggravated by the occurrence in ITER of a quasi-periodic instability (the so-called ELM), which results in a bursty character of the heat release with peak loads exceeding 1 GW_m2 during 1 ms, with typically 10 - 300 Hz repetition rate. The first step in this research project was the development of a PPS for use in the linear plasma generators at DIFFER (Pilot-PSI and Magnum-PSI) which would allow a realistic simulation of the plasma load on the strike zone in ITER, and in particular to superimpose the transient heat/particle load on a the steady state plasma load. A device with that capability did not exist in the world. After that, using this device, the effect of the pulsed load and the combination of pulsed and steady state exposure on the material was investigated. The first goal was achieved by applying current pulses to a running cascaded arc plasma source. This technique, pioneered by Timmermans in 1984, was taken to a new power level which required a number of design changes to the source: a wider plasma channel (to reduce heat load in the source), the cladding of all plasma facing components in the source with molybdenum, the redesign of the cathode and anode and application of a strong longitudinal magnetic field. With those modifications, current pulses of up to 4.3 kA could be superimposed on a steady operation at 200 A, resulting in heat pulses in excess of 1 GW_m2 and an energy density of 0.5 MJ_m2 with millisecond duration, superimposed on steady load of up to 5 MW_m2. At the target, the plasma density reached up to 60_1020m3 at a plasma temperature of up to 12 eV (about 140 thousand K). With those parameters, the PSI conditions that will occur in ITER during ELMs can be realistically simulated. During the source development process, it was discovered that the source plasma purity plays an essential role, leading to an operational bifurcation between two plasma temperature regimes. Without the molybdenum cladding in the source, the plasma was contaminated with copper. Compared to the clean plasma obtained with molybdenum cladding, this resulted in plasmas with much lower temperature (_5 eV instead of _15 eV) but higher electron density (up to 150_1020m3 rather than 60_1020m3). In contrast, the clean hydrogen plasma obtained in the molybdenum-coated source becomes highly ionized, has low radiation losses and goes into the high temperature mode. In order to guide the plasma (pulse) from the source to the target a strong magnetic field was applied (