Material characterization for magnetically confined fusion : Surface analysis and method development

The dream of abundant clean energy has brought scientists and laypeople alike to ponder the possibilities of nuclear fusion since it was established as the energy source of the stars in 1939. Starting from the mid the 20th century, significant effort has been put into overcoming the technological challenges related to the construction of a power plant, but initial optimism has faded somewhat due to a notable absence of practical outcomes. Nevertheless, the research continues and progress is made slowly but surely. The present work deals with a small part of the fusion puzzle, namely the materials to be used in the first wall surrounding a magnetically confined plasma. Carbon, which has historically been considered as the most viable element for this role, has been ruled out due to issues with plasma-induced erosion, hydrocarbon formation and a buildup of thick deposited material layers on wall components. The latter two lead to an unacceptable accumulation of radioactive tritium, both in the deposited layers and in dust particles. A metal wall, which would alleviate these particular problems but increase the severity of others, is therefore envisioned for a future demonstration reactor. Three contributions to the overall research effort are made though this thesis. First, an increased understanding of plasma-induced erosion of so-called reduced activation ferritic-martensitic steels and preferential sputtering of light material components is provided. High-resolution ion beam analysis and microscopy methods are used to examine samples of such a steel after exposure to plasma under controlled circumstances. Model films consisting of a mixture of iron and tungsten deposited on silicon substrates are also studied as they constitute simpler systems where the effects of interest may be simulated. The knowledge obtained is necessary for an assessment of the possibility to use reduced activation steel as a plasma-facing material in specific regions of a reactor wall. The second contribution consists of reports on the composition of deposited material layers on wall components retrieved from the plasma confinement experiments JET and TEXTOR. These provide limited conclusions on the range and rate of material erosion, transport and deposition in two cases. Finally, a detection system for the ion beam technique elastic recoil detection analysis has been assembled, tested and put into operation. In addition to improving the quality of analyses performed on fusion-related materials, the system has become an established tool available for users of the 5 MV electrostatic pelletron accelerator at Uppsala University’s Tandem Laboratory.QC 20190110