Linear plasma devices represent an essential tool for nuclear fusion research, whereby understanding crucial aspects related to plasma-wall interactions or edge plasma behaviour. Simplified models are of great importance to complement and integrate experimental and simulation results of complex systems such as plasmas in linear machines, because they are fast and simple to employ. In this work, we present a global volume-averaged (0D) model for plasma investigation in linear machines. The 0D model equations are based on the space integration of the state of the art edge plasma model implemented in the SOLPS-ITER code. Comparisons between helium plasmas described with 2D simulations performed with SOLPS-ITER and with the 0D model highlight that contributions often neglected in tokamak edge models, e.g. electron-neutral excitation, may be relevant when describing weakly ionized plasmas in linear devices. The model is used to perform sensitivity studies with respect to several parameters and to analyse the time evolution of the system, leading to the identification of two relevant time scales governing the system. Lastly, a comparison of 0D results with experimental data from the linear device GyM is performed, showing satisfactory agreement. Our methods and results provide crucial interpretative keys in the investigation of the physics of edge plasmas.

Plasma–material interaction (PMI) in tokamaks determines the life-time of first-wall (FW) components. Due to PMI, FW materials are eroded and transported within the device. Erosion is strongly influenced by the original morphology of the component, due to particle redeposition on near surface structures and to the changing of impact angle distributions, which results in an alteration of the sputtering effects. The Monte-Carlo impurity transport code ERO2.0 is capable of modelling the erosion of non-trivial surface morphologies due to plasma irradiation. The surface morphology module was validated against experimental data with satisfactory agreement. In this work, we further progress in the validation of the ERO2.0 capabilities by modelling both numerically generated surfaces as well as real surfaces, generated using atomic force microscopy (AFM) measurements of reference tungsten samples. The former are used to validate ERO2.0 against one of the morphology evolution models present in literature, in order to outline the conditions for reliable code solutions. Modifications induced in AFM-generated surfaces after argon and helium plasma irradiation are compared, showing a similar post-exposure morphology, mostly dominated by surface smoothing. Finally, the ERO2.0 morphology retrieved after He plasma exposure is compared to experimentally-available scanning electron microscopy and AFM measurements of the same surface morphology exposed in the linear plasma device GyM, showing the need for further improvements of the code, while a good agreement between experimental and simulated erosion rate is observed.

In order to produce the 195m' 9Hg/195mA~ genera tor for myocardial. studies (1,2 and references therein,3), the possibility to use on-line thermal diffusion/evaporation method was speculated (1,4). In fact, the Au/Hg separation is possible by thermal diffusion o f radio-Hg through thin Au sheets (either in vacuum or with an inert gas as a carrier) in the temperature range from 700 to 800 OC, b elow the gold melting point (1,4). A gold multitarget (GMT) made by 15 gold sheets, 50 pm thick (i.e: 0.75 mm Au = 3 3 . 3 -> 20 MeV proton energy loss ( 5 ) ) , distance<] from 3 to 7 mm one to the other, seems t o be suitable for on-lin(, separation of radio-H9 from gold using the proton beam power i tself (13.3 W/pA) for heating the system.In order to make a theoretical evaluation of the temperature profile Ti(x,y) induced by the proton beam on the i-th target, a computer code named HEAT was developed: ration: 1. -radial heat conduction diation (Stefan-Boltzmann law) and 3 cooling fluids (Newton law), as a profile Ji(x,y) on the i-th metallic Obviously the i-th temperature prof this code takes into conside-. 1-st Fourier law), 2. -irra--convection with carrier o r function of the incident beam sheet. le depends strongly on the J i (x,y) beam profile, whose shape depends strongly on the interaction with the previous (i-1)-th metallic sheet, as well as on the shape of the primary incident beam, J1(x,y), and the distances between and the thicknesses of the various sheets.S o , in order to evaluate the effect of multiscattering and plurime scattering ( 6 , 7 ) on the 15 gold sheets of the GMT, a Monte Carlo computer code named was developed (4). The code !ST can use the code HEAT as a subroutine.Any shape of the incident beam profile J1(x,y) can be used (gaus-

Plasma-wall interaction (PWI) is a great challenge in the development of a nuclear fusion power plant. To investigate phenomena like erosion of plasma-facing components, impurity transport and redeposition, one needs reliable numerical tools for the description of both the plasma and the material evolution. The development of such tools is essential to guide the design and interpretation of experiments in present and future fusion devices. This contribution presents the first global simulation of PWI processes in a linear plasma device mimicking the boundary plasma conditions in toroidal ones, including both the description of plasma and impurity transport and of plasma-facing material evolution. This integrated description is obtained by coupling two of the state-of-the-art numerical codes employed to model the plasma boundary and the PWI, namely SOLPS-ITER and ERO2.0. Investigation of helium plasma is also of primary importance due to the role helium will have during ITER Pre-Fusion Power Operation, when it is planned to be used as one of the main plasma species, as well as fusion ash in Full Power Operation. The plasma background is simulated by SOLPS-ITER and the set of atomic reactions for helium plasmas is updated, including charge-exchange and radiative heat losses. ERO2.0 is used to assess the surface erosion in the GyM vessel, using different wall materials (e.g. carbon, iron or tungsten) and applying different biasing voltage. Eroded particles are followed within the plasma to assess their redeposition location. The ionization probability of the different materials in the GyM plasma is inferred through the energy distribution of impacting particles and its effects on migration are investigated.

No abstract