Plasma Window Performance Leslie Bromberg Katie Maurer Ady

Plasma Window Performance Leslie Bromberg Katie Maurer Ady Herskovitch* MIT Plasma Science and Fusion Center ARIES meeting Madison, WI April 23, 2002 *Brookhaven National Laboratory, NY

HIBD-Chamber Vacuum Interface • Heavy Ion Beam Driver requires high vacuum for operation • 10 -6 -10 -9 Torr • Chamber operation requires low to intermediate vacuum • 10 -3 - 10 Torr • Because of the large openings required for beam propagation, large gas throughput across the HIB final focus and the chamber exits • Large vacuum pumping speeds required • Not clear whether it is possible to maintain that large pressure differential with the available space for pumps. • Can Flinabe pump Xe?

Plasma Window • Experiments at MIT for supporting work on a compact neutron source (Dr. R. Lanza, Dr. A. Herskovitch) • Funded in the past through LLNL

Schematic of plasma window operation

Argon viscosity at 1 bar

Throughput calculations • In the viscous regime (usually p > 100 m. Torr), the throughput through a channel is Here, Q is the throughput, h is the gas viscosity, a is the diameter, and P’s are the pressure • Increased viscosity and decreased number density results in decreased flow through the opening. • If the channel is filled with a thermal plasma, both the viscosity increases and the number density decreases, decreasing the particle throughput.

Plasma window picture as of 1/8/2002

Preliminary results in Argon diameter of orifice: 6 mm

Plasma windows for applications to HID • Demonstrated technology for intermediate pressure (in the viscous regime) • • • Preliminary experiments with Ar at atmospheric and higher pressures Minimum pressure at low pressure side is ~ 30 m. Torr (down from 100 m. Torr) Not clear how low it can reach with different gases, different size openings and lowered pressure at high pressure side of window • Need scaling experiments • Additional experiments at lower pressures and with other gases to come in the summer • Power consumption, per window, is probably (? ) on the order of 50 -200 W for 1 Torr operation with Xe • Requires dielectrics (? ) • Already a conducting media