Particle Control and Transport Experiments in the DIII-D Tokamak with Graphite Walls

Full text: Retention of hydrogenic isotopes in the first wall of future burning plasma experiments such as ITER can lead to an unacceptable in-vessel tritium inventory. To address this topic, we report new results from recent particle control and transport experiments in the DIII D tokamak. We find that dynamic particle balance calculations (particle sources and sinks calculated vs time) yield similar results to a shot-averaged 'static' (calculated by pressure rise) particle balance. The dynamic particle balance measurements show very low wall retention in H-mode, compared with large retention in L-mode; ELMing H-mode discharges with either NBI or EC heating show similar results. Particle balance in discharges with resonant magnetic perturbation (RMP) ELM suppression show that the wall retention rate and inventory is dependent on pedestal density and divertor conditions. The dominant term in the particle balance equation during RMP discharges was the cyro-pump exhaust rate. 13C injection experiments have shown that most of the carbon is deposited at the inner strike point of a SN divertor in L-mode, additional deposition in the private flux zone is present in H-mode. With an unbalanced DN plasma shape, there is more localized 13C deposition near the injection point in the non-active divertor. Moderate heating of graphite DiMES samples to 200oC results in a factor of 10 less deposition compared to room temperature. A new Porous Plug Injector has been used to demonstrate that chemical erosion in the graphite divertor target is dramatically reduced in a cold, detached divertor. Preparations for a DIII-D demonstration of removal of re-deposited carbon with an oxygen bake have progressed and final reviews are in progress. An oxygen bake of 2 Torr for 2 hours at 350oC is predicted to remove deposited 13C layers in DIII-D. In ITER, removal of the carbon would also remove the co-deposited tritium. A new technique involving processing previously exposed and analyzed DIII-D tiles 'known to be rich in 13C' would be processed in DIII-D. These could be installed during a clean vent. This tests 13C removal in 'remote' regions of the torus, and provides a better test of restart of plasma operations after the oxygen bake. This work was supported by the US Department of Energy under DE-AC52-07NA27344 and DE-FC02-04ER54698. (author)