Effects of collisionality and T e /T i on fluctuations in positive and negative δ tokamak plasmas

The effects of negative triangularity () on confinement and fluctuations in plasmas covering a large range of parameters were investigated on the tokamak à configuration variable (TCV). The conditions explored in this paper include discharges where neutral beam (NB) heating was employed to obtain an electron-ion temperature ratio across a large fraction of the plasma profile. This significantly extended the range of negative plasmas studied on TCV towards conditions more relevant to future reactor-like tokamaks. Negative triangularity was found to improve confinement over the full range of collisionality studied () and (). The amplitude of radiative temperature fluctuations, measured using a correlation electron cyclotron emission diagnostic over the range , was found to be reduced, in negative with respect to positive plasmas, for all combinations of parameters explored. This was, in particular, verified for a pair of positive and negative plasmas with comparable density and under different conditions of NB heating. Linear gyrokinetic simulations found the dominant turbulence regime, in the strongly NB heated discharges, to be a mixture of trapped electron modes (TEMs) and ion temperature gradient driven modes. This is in contrast to ohmic or electron cyclotron heated discharges, for which the dominant turbulence regime was found to be pure TEM. Negative triangularity was found to lead to partial stabilization of the most unstable modes for low wavenumbers in both turbulence regimes. These findings demonstrate that negative triangularity could provide significant confinement improvement over a large range of parameters, that include conditions closer to future reactor-like machines (, low collisionality).


Publié dans:
Nuclear Fusion, 60, 1
Année
Oct 30 2019
Mots-clefs:
Laboratoires:




 Notice créée le 2020-01-10, modifiée le 2020-01-13

PREPRINT:
Télécharger le document
PDF

Évaluer ce document:

Rate this document:
1
2
3
 
(Pas encore évalué)