Pedestal and core confinement of hybrid scenario in ASDEX Upgrade and DIII-D

Pedestal and core confinement of hybrid discharges in ASDEX Upgrade (AUG) and DIII-D are studied in dedicated power scan experiments. The H98(y, 2) confinement factor increases with total beta(N) in both tokamaks and it is higher in DIII-D with higher delta plasma shape at a given beta(N). The pedestal beta, beta(PED)(N), increases linearly with total beta in AUG hybrid discharges, while it is roughly constant with beta(N) at fixed shape in the DIII-D power scans. The confinement enhancement with power observed with respect to the IPB98(y, 2) scaling is due to an increase in pedestal confinement in AUG hybrid discharges and to an increase in core confinement in the DIII-D hybrid power scans. The increase in pedestal pressure with power in AUG hybrid discharges is primarily due to an increase in the width of the edge transport barrier at constant pressure gradient. In the DIII-D discharges the widths of the T-e and n(e) pedestals, Delta(Te) and Delta(ne), are consistent with a Delta similar to beta(1/2)(pol,PED) scaling. In the AUG hybrid power scans a dependence of Delta(Te) on beta(pol,PED) cannot be excluded, while Delta(ne) shows no dependence on beta(pol,PED) In both machines Delta(Ti) increases with beta. The maximum pedestal pressure achieved in the experiment prior to the onset of type I ELMs is consistent with predictions from ideal MHD; however, a physics model explaining the increase in the pedestal width with beta is still missing. The increase in R/L-Ti with beta in the core of DIII-D is consistent with predictions by linear gyrokinetic simulations. In the plasma core, E x B shearing rate stabilization of the ITG modes is significant in both machines as beta is increased. Inclusion of electromagnetic effects in the gyrokinetic calculations provides additional stabilization at beta(N) values achieved in the experiment. In AUG, proximity to the kinetic ballooning threshold and/or a stronger reduction in normalized ion heat flux with increasing input power are possible explanations for the constancy of R/L-Ti at mid-radius as beta is increased.