Description
The effect of the density and collisionality on the intrinsic rotation of Ohmic plasmas at JET was studied experimentally in pure Hydrogen, pure Deuterium and pure Tritium plasmas. Two rotation reversals were observed for each of the hydrogen isotopes, with rotation profiles at mid-radius changing from peaked to hollow at a density close to the transition from the linear to the saturated Ohmic confinement regimes, then to peaked again with increasing density [1]. Modeling of main-ion rotation profiles measured at different densities, was performed with a version of the multi-scale gyrokinetic code GS2 [2] where low-flow gyrokinetic equations, as described in [3], have been implemented [4]. Here, we focus on one of the turbulence drives, the effect on turbulence of the neoclassical parallel velocity and heat flow, and the neoclassical poloidal electric field. GS2 was coupled with the neoclassical transport code NEO [5] used to calculate corrections to the background neoclassical distribution functions. Since GS2 is a local flux-tube code, momentum fluxes were obtained from GS2 nonlinear runs at several radial positions. The separate contributions of diffusion and Pinch terms and, Πintrinsic (defined as the toroidal angular momentum flux in the absence of flow and flow shear) were calculated. Modelling was performed first for one Deuterium discharge with a hollow rotation profile, where the core of the plasma rotates in the opposite direction to the outer region. This confirmed that the model was able to produce hollow rotation profiles with core counter-rotation and momentum fluxes of the right order of magnitude to explain the measured rotation angular frequency. Modelling was also done for discharges from a density scan of Hydrogen plasmas. Observations show that as the density increased, rotation profiles went from peaked to hollow and then to peaked again. The GS2 simulations reproduce both the low-density and the high-density reversals. The model and the experimental data agree in the sign of the velocity gradient, however for peaked profiles the model predicts velocity gradients smaller than those in the experiment.
References:
[1] M.F.F. Nave et al. Nucl. Fusion 63, 044002 (2023)
[2] W. Dorland et al, Phys. Rev. Lett. 85, 5579 (2000)
[3] F.I. Parra and M. Barnes, Plasma Phys. Control. Fusion 045002 (2014)
[4] M. Barnes et al, PRL 111, 055005 (2013).
[5] E. A. Belli and J. Candy, Plasma Phys. Control. Fusion 54.1 (2012).