Evolution of the H-mode edge pedestal between ELMs

Abstract

The evolution of edge pedestal parameters between edge-localized modes (ELMs) is analyzed for an H-mode DIII-D (Luxon 2002 Nucl. Fusion 42 612) discharge. Experimental data are averaged over the same sub-intervals between successive ELMs to develop data that characterize the evolution of density, temperature, rotation velocities, etc over the interval between ELMs. These data are interpreted within the context of the constraints imposed by particle, momentum and energy balance, in particular in terms of the pinch-diffusion relation for radial particle flux that is required by momentum balance. It is found that in the edge pedestal there is an increase in both inward (pinch) electromagnetic and outward (diffusive) pressure gradient forces over the inter-ELM interval.

The effect of ion orbit loss and X-loss on the interpretation of ion energy and particle transport in the DIII-D edge plasma

“…The minimum initial energy required for an ion with direction cosine f 0 located at poloidal position h 0 on flux surface "0" to be able to cross the last closed flux surface "D" (at r D ¼ a) at poloidal location h D can be calculated from Eq. (5). The smallest such energy for all values of h D is the minimum energy required for an ion with direction cosine f 0 located at poloidal position h 0 on flux surface "0" to be able to cross the last closed flux surface "D" (at r D ¼ a).…”

“…A relationship between changes in the radial electric field E r and in the poloidal rotation velocity V h in the plasma edge and changes in the edge pressure, temperature, and density gradients in the plasma edge has long been observed experimentally, 3 suggesting that an understanding of the causes of the rotation velocities and the radial electric field may provide insight to an understanding of edge pedestal physics, and recently it has been demonstrated that changes in these experimentally observed quantities are correlated by momentum balance requirements. [4][5][6] A second, and more widely held, school of thought postulates that the stabilization or destabilization of electromagnetic microinstabilities [7][8][9] and the corresponding changes in fluctuation-driven transport cause the observed changes in temperature and density gradients in order for diffusive heat and particle fluxes to remove the input heat and particles. A third school of thought is that the physics of the edge plasma is determined, at least in part, by the loss of energetic ions and their consequences.…”

X-transport of ions in diverted tokamaks, with application to DIII-D

“…Implicit in much of this fluid code analysis is the assumption that the particle and heat fluxes in the edge plasma are diffusive in nature (i.e., that the particle and heat fluxes are proportional to gradients of the measured density and temperature profiles), although many of these codes have the capability to represent convective transport and some recent analyses have explicitly taken into account the particle pinch required to satisfy momentum balance. [10][11][12] However, direct ion-orbit-loss (i.e., the loss of particles and energy due to particles following orbits which leave the confined plasma and do not return) is not usually taken into account in interpreting edge pedestal physics. In the "standard" type of ion-orbit-loss (e.g., as described in Ref.…”

The role of radial particle pinches in ELM suppression by resonant magnetic perturbations

“…The lead author and his colleagues have made numerous calculations of carbon and deuterium poloidal velocities over the years for models of DIII-D and other plasmas (e.g. [27][28][29][30]). Two general conclusions emerge from these calculations.…”

“…Two general conclusions emerge from these calculations. (1) In both the edge and the core, when the friction force is much less than the parallel viscous force, the deuterium and carbon ions rotate in opposite directions with different magnitudes, but when the friction force is much greater than the viscous force then both species rotate in the same direction with similar magnitudes (calculations reporting such results are given in [27][28][29][30]). (2) In the plasma edge, the measured carbon rotation velocity is predicted (by the models described in [27] and the references therein) reasonably well in the flattop region, but is significantly over-predicted in the steep gradient region, probably indicating that a retarding torque (ion orbit loss?…”