Analysis of metallic impurity density profiles in low collisionality Joint European Torus H-mode and L-mode plasmas

This paper describes the behavior of nickel in low confinement (L-mode) and high confinement (H-mode) Joint European Torus (JET) discharges [P. J. Lomas, Plasma Phys. Control. Fusion 31, 1481 (1989)] characterized by the application of radio-frequency (rf) power heating and featuring ITER (International Thermonuclear Experimental Reactor) relevant collisionality. The impurity transport is analyzed on the basis of perturbative experiments (laser blow off injection) and is compared with electron heat and deuterium transport. In the JET plasmas analyzed here, ion cyclotron resonance heating (ICRH) is applied either in mode conversion (MC) to heat the electrons or in minority heating (MH) to heat the ions. The two heating schemes have systematically different effects on nickel transport, yielding flat or slightly hollow nickel density profiles in the case of ICRH in MC and peaked nickel density profiles in the case of rf applied in MH. Accordingly, both diffusion coefficients and pinch velocities of nickel are found to be systematically different. Linear gyrokinetic calculations by means of the code GS2 [M. Kotschenreuther, G. Rewoldt, and W.M. Tang, Comput. Phys. Commun. 88, 128 (1995)] provide a possible explanation of such different behavior by exploring the effects produced by the different microinstabilities present in these plasmas. In particular, trapped electron modes driven by the stronger electron temperature gradients measured in the MC cases, although subdominant, produce a contribution to the impurity pinch directed outwards that is qualitatively in agreement with the pinch reversal found in the experiment. Particle and heat diffusivities appear to be decoupled in MH shots, with χe and DD≫DNi, and are instead quite similar in the MC ones. In the latter case, nickel transport appears to be driven by the same turbulence that drives the electron heat transport and is sensitive to the value of the electron temperature gradient length. These findings give ground to the idea that in ITER it should be possible to find conditions in which the risk of accumulation of metals such as nickel can be contained.