Stability of snowflake Diverted and Negative Triangularity Plasmas in the TCV Tokamak

The TCV experimental campaign aims at studying H-mode plasmas and ELM behavior in various diverted configurations including equilibria with negative triangularity [1] and snowflake divertor. Although the TCV tokamak is capable of generating plasma cross sections with a large variety of shapes, certain combinations are excluded by technical constraints. In particular, for diverted configurations, the strike points should be placed in regions covered by graphite tiles, which can support the local power densities. A second order null divertor (snowflake) has been successfully created and controlled in the TCV tokamak [2]. Variations of the edge stability for the snowflake equilibria at different values of triangularity are investigated. 1 Stability of negative triangularity equilibria The values of the n = 0 growth rates are very high for the negative (both upper and lower) triangularity equilibria in the TCV tokamak presented in [1] with a reference elongation κ = 1.75: � 1500s 1 . The growth rate is further increased when taking into account the ports (modeled as very high resistivity axisymmetric pieces of the wall): � 3000s 1 in the reference plasma position and even can go ideally unstable when the plasma is shifted inside farther from the LFS wall. It confirms the conclusion about the high sensitivity of vertical stability for negative triangularity configurations to the distance between the outer wall and the plasma. A lower elongation is needed for negative triangularity configurations to be vertically controllable. Growth rates of axisymmetric modes were computed for fixed boundary equilibrium series with negative triangularity and elongation κ = (κup + κdown)/2 varied from 1.4 to 1.75. The up-down non-symmetric plasma boundary was prescribed keeping the ratio κup/κdown = 0.75 and with triangularity δup = δdown = 0.65. The dependence of the n = 0 growth rates is close to linear in (κ 1) and the growth rate value doubles when κ varied from 1.4 to 1.75. The elongation κ = 1.4 corresponds to a reasonable growth rate � 700 s 1 without ports (a factor of 2 larger with axisymmetric openings in the place of ports). The upward shifted (away from the LFS port) plasma is more vertically stable, however the issues related to the separatrix strike point positions at the LFS wall are to be solved in this case. With an elongation increased up to 1.5 1.6, the divertor strike points can be placed in full toroidal coverage zones with graphite tiles at the LFS wall. The SPIDER free boundary equilibrium code was used to compute series of equilibria using the control plasma shape with elongations κup = 1.2, κdown = 1.8; δdown = 0.65 and varying δup to be matched by the ψ/ψsx =