Predictive modelling of the transport of impurities emitted from a liquid metal divertor in the core of the EU-DEMO future fusion reactor

The aim of this thesis is the development of an integrated model for the transport of impurities ejected from the surface of a liquid metal divertor (LMD) inside the core plasma of the European DEMO (EU-DEMO) future fusion reactor. Nuclear fusion is a promising energy production system based on the process that lights up the stars, and the great challenge is to reproduce suitable conditions for fusion reactions to occur on Earth. The magnetic confinement approach, realized in machines called Tokamaks, consists in confining the fusion fuel - a plasma heated up to ∼ 100 million Celsius degrees - through a system of magnets. One of the key challenges to be solved is represented by the interactions between plasma and wall materials, where a fraction of the total power must be exhausted. This role is fulfilled by the divertor, where most of the heat and particles are directed and where very effective cooling mechanisms are employed. In view of the significantly larger amount of power produced by EU-DEMO (the first tokamak planned to produce electricity that was chosen as the reference system for this work) with respect to present-day experiments, the implementation of a self-healing liquid metal divertor, unlike traditional solid one, has been considered. This solution can exploit the evaporation process resulting from the heated plasma impacting its surface, leading to the generation of a cloud of particles in front of the divertor plates, which contribute to exhausting the incoming power via radiation processes. However, should the impurities evaporated from the wall reach the core plasma, the resulting pollution could contaminate it to the point that the fusion process is hindered. Studying the transport of these particles which can cause the plasma shutdown, throughout the Scrape Off Layer (SOL) - a thin region surrounding the reacting fusion plasma core - and the core itself, is necessary to assess the feasibility of the liquid metal divertor option, thus representing a strong motivation for this thesis, which can be regarded as a first step towards an integrated divertor-SOL-core model. As the initial phase of the work, transport phenomena due to classical, neoclassical, and turbulent effects have been thoroughly reviewed, to study the motion of impurities inside core plasma. Afterward, a comparison between Lithium and Tin (the most promising materials for a liquid metal divertor) in terms of their impact on the core plasma power balance has been performed. This required the study of their (charge-state-wise) radial density distribution in the core plasma, identifying possible phenomena leading to undesired accumulation. This calculation has been performed using the ASTRA and STRAHL codes, capable of solving respectively the transport equation of background plasma and of impurities at different ionization stages. Neoclassical and turbulent effects have been considered in evaluating transport coefficients by means of NCLASS and TGLF codes, eventually leading to the calculation of the radial profiles of impurities and the correspondent radiated power. After having validated some preliminary results against shots from ASDEX-Upgrade, simulations for the EU-DEMO scenario have been prepared, both for the case of Lithium and Tin. Promising results, in particular for Lithium, seem to describe a shift of impurities to the outer region of the core, avoiding central pollution and confining radiations on the edge, where they are beneficial to decrease the wall temperature.