Recent H-mode density limit experiments on DIII-D

摘要

A vast body of tokamak data is in good agreement with the empirical density limit scalings proposed by Hugill and Greenwald. These scalings have common puzzling features of showing no dependence on either impurity concentration or heating power, since the density limit is frequently correlated with a rapid rise of the edge radiation. Despite the resiliency of these scalings, several machines under restrictive conditions have operated at densities well above the predictions of these scalings, albeit with pellet injection and at varying degrees of confinement degradation. Furthermore, data from several machines display a weak dependence on heating power. These results cast doubt on the universal validity of both of these scalings. Nevertheless the fact remains that access to densities above Hugill-Greenwald scaling is very difficult. A number of theories based on radiative power balance in the plasma boundary have explained some but not all features of tokamak density limit behavior, and as ITER design studies recently brought to focus, a satisfactory understanding of this phenomenon is lacking. Motivated by a need for better understanding of effects of density and fueling on tokamak plasmas in general, the authors have conducted a series of experiments designed to identify and isolate physical effects suspected to be directly or indirectly responsible for the density limit. The physical effects being considered include: divertor power balance, MARFE, poloidally symmetric radiative instabilities, MHD instabilities, and transport. In this paper they first present a brief summary of the experimental results up to the writing of this paper. The remainder of the paper is devoted to a comparison of this data at the onset of the MARFE instability with predictions of theory and the implications of the results on access to densities beyond the Hugill-Greenwald limit.