Study of positive ion transport to the plasma electrode in giant RF negative ion sources

Negative ion sources are fundamental components of neutral beam injectors (NBI), one of the main heating systems for fusion reactors. SPIDER is the full-scale prototype negative ion source for ITER NBIs. It is hosted in Padua as part of the Neutral Beam Test Facility (NBTF). It aims to extract up to 330 A m-2 of negative hydrogen ions from an inductively coupled plasma, generated inside 8 cylindrical drivers. The negative ion production is enhanced by caesium evaporation inside the source. In caesium-seeded negative ion sources, negative ions are produced close to the extraction apertures, and they are mainly generated by surface conversion of neutral atoms and positive ions impinging on the ion source walls, particularly on the plasma grid. The conversion yields depend on the energy distribution of these precursors, and so does the energy of those particles which are reflected as negative ions. The positive ion flow in the extraction region may also impact on the extraction probability of negative ions, via momentum transfer. Besides, in giant multi-driver RF sources such as SPIDER, a gradient of plasma potential is present in the expansion region Sartori et al. (2021), affecting the positive ion transport towards the caesiated plasma electrode and their energy. To approach this complex problem, a 3D test-particle Monte Carlo code for tracing plasma motion in SPIDER was developed. Positive ions species are generated in different positions within the plasma source volume and are tracked under the influence of electric and magnetic fields. Then, Monte Carlo collisions are used to simulate the interaction with predetermined backgrounds of plasma and neutrals, with profiles derived from experimental data. The particles are traced until they hit the ion source walls. Finally, the energy distribution of the different particle species impinging on the plasma grid (PG) are determined, and used to assess the generation and the energy distribution of the produced H.