Experimental studies of Alfven mode stability in the JET tokamak

Full text: Controlling the interaction between alpha's and modes in the Alfven frequency range is a crucial issue for ITER operation in order to prevent alpha's losses. We study experimentally the dependence of the stability threshold of Alfven Eigenmodes (AEs) with various toroidal mode numbers (n's) on different Ion Cyclotron Resonance Frequency (ICRF) heating schemes and background plasma parameters. Multifrequency ICRF heating gives rise to a lower perpendicular fast ion temperature in the plasma core. By spreading the power deposition profile, this lowers the fast ion pressure gradient, hence leading to a reduced AE activity. The use of 2nd harmonic minority ICRF heating at higher plasma density reduces the fast ion temperature in the plasma core, and gives rise to a lower fast ion driven plasma rotation. A controlled radial redistribution of the alpha's can be beneficial for the stability of AEs with intermediate n's. This mechanism is experimentally simulated in JET by using the saddle coils to produce error fields that couple to the q=2 surface. This scheme gives rise to a significantly higher threshold for core localised AEs in the presence of error fields, which we attribute to a modification of the radial profile of the fast ion distribution function. The direction of the ion gradB-drift is an important parameter in determining the accessibility conditions for the H-mode regime, with multi-machine experimental evidence that a lower power is required for the L-H mode transition when the ion gradB-drift is directed towards the divertor. We find that the damping rate of n=1 TAEs with similar frequency and radial location is approximately a factor three higher for the case of ion gradB-drift directed away from the divertor. The present JET saddle coil system can only excite low-n AEs, whereas the n's that can be driven unstable in ITER by the alphas are expected to be in the range n∼5- 20. The mismatch between the modes driven by the saddle coils and those that are made unstable by the fast particles is already observed on JET. Building on the present system, a new set of antennas to drive AEs with mode numbers up to n∼10-15 is being designed for JET. Design principles and constraints will be presented along with the results of the coupling and engineering analysis, together with a discussion of the possible extrapolation of such a system to ITER.(author)