Numerical Investigation of Vapour Shielding in Linear Plasma Devices

Nuclear fusion technologies are one of the most investigated lines of research to try to tackle the problem of finding a reliable, sustainable and environmental-friendly energy source. Notwithstanding all the efforts that have been put into it, there are still several issues that hinder its development: one of them is the interaction between the plasma and the plasma-facing components, because of the high energy and particle flux that the target surfaces are requested to bear. The currently adopted divertor technologies are still far from being able to withstand the fluxes that a future plant could supply and it is thus needed a design revolution. A breakthrough in this respect could consist in the use of divertors coated by liquid metals (LMs) instead of the bare solid ones: they exhibit a self-healing nature and they are able to produce a vapour cloud in front of the target, which is able to dissipate a relevant fraction of the energy before it can reach the surface. Experiments performed exploiting linear plasma devices (LPDs) showed that the physical parameters of the vapour cloud have an oscillating dynamical behaviour, which is induced by a detachment-like phenomenon of the plasma and by the difference of timescales between thermal equilibria of the LM and the atomic physics. This thesis aims at finding a suitable zero-dimensional model to study the interaction between the plasma and the liquid metal in LPDs. The use of such a simplified model can be justified by the fact that it can be fast-running (since it focuses on the time dependence of the relevant phenomena) and that it is able to include terms that would be otherwise neglected in much more complex codes, possibly allowing additional interpretations of the phenomena at play. Taking as a starting point a 0D model that referred only to the main plasma species and to the whole volume of a LPD, it was advanced in order to include the LM populations. The studies performed with such a model showed that employing input parameters resembling more physical situations (in which the average LM density over the whole device is small if compared to the main plasma), the model is appropriate but it is not able to show a significant plasma-LM interaction. It is instead possible to do so by increasing the LM concentration, in order to simulate the vapour cloud situation in which the main plasma and LM densities are comparable: in this case an important synergy between the two species has been noticed, but it was not possible to study the phenomena in a self-consistent way. To solve both these issues it will be needed in the future to develop a 0D model whose domain is readjusted to the volume of the vapor cloud in front of the divertor: the set of equations that could give birth to such a model are discussed at the end of this work.