Techniques For Injection Of Pre-Charaterized Dust Into The Scrape Off Layer Of Fusion Plasma
Authors: A.L. Roquemore, B. John, F. Friesen, K. Hartzfeld, D.K. Mansfield
Abstract: Introduction of micron-sized dust into the scrape-off layer (SOL) of a plasma has recently found many applications aimed primarily at determining dust behavior in future fusion reactors. The dust particles are typically composed of materials intrinsic to a fusion reactor. On DIII-D and TEXTOR [1] carbon dust has been introduced into the SOL using a probe inserted from below into the divertor region. On NSTX, both Li and tungsten dust have been dropped from the top of the machine into the SOL throughout the duration of a discharge, by utilizing a vibrating piezoelectric based particle dropper [2]. The original particle dropper was developed to inject passivated Li powder ~ 40 μm in diameter into the SOL to enhance plasma performance. A simplified version of the dropper was developed to introduce trace amounts of tungsten powder for only a few discharges, thus not requiring a large powder reservoir. The particles emit visible light from plasma interactions and can be tracked by either spectroscopic means [3] or by fast frame rate visible cameras [4]. This data can then be compared with dust transport codes such as DUSTT [5] to make predictions of dust behavior in next-step devices such as ITER. For complete modeling results, it is desired to be able to inject pre-characterized dust particles in the SOL at various known poloidal locations, including near the vessel midplane. Purely mechanical methods of injecting particles are presently being studied using a modified piezoelectric-based powder dropper as a particle source and one of several piezo-based transducers to deflect the particles into the SOL. Vibrating piezo fans operating at 60 Hz with a deflection of ± 2.5 cm can impart a significant horizontal boost in velocity. The highest injection velocities are expected from rotating paddle wheels capable of injecting particles at 10’s of meters per second depending primarily on the rotation velocity and diameter of the wheel. Several injection concepts have been tested and will be discussed below.
__________________________________________________
Submitted to: Fusion Engineering and Design (Sept. 2010)
__________________________________________________
Download PPPL-4643 (pdf 13.7 MB 4 pp)
__________________________________________________