L-H transition studies on spherical tokamaks

While the higher fusion power achieved in high confinement (H-)mode make it a desirable regime for tokamak devices, the transition from low confinement (L-)mode to H-mode is a complex, multi-faceted process whose dynamics are still not fully understood. It would not be possible to explore the entire physics of the L-H transition in a single project, but a selection of aspects have been studied and presented here for the spherical tokamaks MAST and MAST-U. Spherical tokamaks are known to differ from conventional tokamaks in a number of physics areas, including the L-H transition, but our understanding is limited by a comparative lack of experimental data. A comprehensive study of the density dependence of the H-mode power threshold P LH on MAST is presented, showing for the first time the presence of the low-density branch and describing different types of boundary behaviours and intermediate confinement states. As expected for spherical tokamaks, the commonly used empirical scaling laws significantly underestimate P LH . The low-density branch is further explored in a comprehensive study of the L-H transition values of the ion and electron heat flux in the edge. Here the effects of fast ion losses on the NBI heating efficiency are demonstrated, and a strong density dependence of the absorbed NBI power is found, independent of the actual injected power. The presence of the low-density branch of P LH is not explained by the heating effects however, and appears to originate in the rate of change of stored energy instead. As the improved confinement in H-mode is characterised by a suppression in turbulence, the turbulence data in the form of 2D density fluctuations measured by beam emission spectroscopy (BES) is explored for selected L-H transition shots, with a cross-correlation time delay estimation (CCTDE) technique used to calculate poloidal and radial velocities from BES data. Finally, a set of experiments performed on MAST-U to explore the power threshold in conventional and Super-X divertor configurations is described and preliminary results are presented.