'Lawrence Livemore National Laboratory, Livemore, CA, USA "General Atomics, P.O. Box 85608, San Diego, CA 92186-9784, USA Pellet injection has been used on the DIII-D tokamak to study density limits and particle transport in H-mode and inner wall limited L-mode plasmas. These experiments have provided a variety of conditions in which to examine the fueling efficiency of pellets injected into DIII-D plasmas. The fueling efficiency defined as the total increase in number of plasma electrons divided by the number of pellet fuel atoms, is determined by measurements of density profiles before and just after pellet injection. We have found that there is a decrease in the pellet fueling efficiency with increased neutral beam injection power. The pellet penetration depth also decreases with increased neutral beam injection power so that, in general, fueling efficiency increases with penetration depth. The fueling efficiency is generally 25% lower in ELMing H-mode discharges than in L-mode due to an expulsion of particles with a pellet triggered ELM. A comparison with fueling efficiency data from other tokamaks shows similar behavior.
IntroductionPellet fueling is an important technique developed for fueling and density profile control in a fusion grade plasma [l]. Much effort has been devoted in past pellet fueling experiments to understand pellet ablation and pellet induced changes in plasma transport.These studies have yielded an extensive validation of the neutral gas shielding (NGS) scaling law for pellet penetration [2] and have shown the capability of pellet fueling to strongly modify the density profile shape. device (an elongated, diverted, tokamak plasma) under various operating conditions and compare these experiments to the overall database results. The fueling efficiency, which is defined as the total increase in number of plasma electrons divided by the number of fuel atoms in the pellet, is determined by Thomson scattering measurements of density profiles before and just after pellet injection in conjunction with the pellet mass measured in a microwave cavity. These density measurements have been made as close as 20 ps after completion of the pellet ablation process, but are more typically made 1 to 2 ms after injection of the pellet. Multiple measurements made starting 150 ps following ablation show only very modest changes in the density profile in the first 2 ms following injection [7] after the initial rapid response to the pellet deposited mass. This indicates that no fast particle transport effects occur during this time.
Pellet plasma interactionThere are several effects that can reduce pellet fueling efficiency from the ideal 100% level. First, there is some ablation of the pellet in the scrape off layer (SOL) before it reaches the last closed flux surface or seperatrix of the plasma. This is due to energetic particles in the SOL that impinge on the pellet as it traverses the SOL. The magnitude of this ablation is rather small on DIII-D as measured by D, light emitted by the ablating ...