| Presentation: | submitted: | by: |
|---|---|---|
| icctalk.pdf | 2007-02-15 10:07:51 | John Berkery |
Confinement of Pure Electron Plasmas in the Columbia Non-neutral Torus
Author: John W Berkery
Requested Type: Consider for Invited
Submitted: 2006-12-15 14:19:40
Co-authors: J.P. Kremer, T. Sunn Pedersen, R.G. Lefrancois, Q.R. Marksteiner, M.S. Hahn
Contact Info:
Columbia University
30 River Court, Apt. 2608
Jersey City, NJ 07310
USA
Abstract Text:
Small Debye length, low temperature, pure electron plasmas have been
successfully confined in the Columbia Non-neutral Torus (CNT). These
plasmas are created by steady state emission of electrons from an
electron emitter physically located on the axis of the magnetic
surfaces. Equilibrium density, temperature, and potential profiles
have been measured by emissive Langmuir probes and they agree well
with numerical calculations.
Confinement time is measured by dividing the total number of
electrons by the emission current, which is equal to the electron
loss rate. The confinement times are found to be very long compared
to the relevant time scales for parallel force balance and
macroscopic stability. This indicates that direct bad orbit losses
are negligible and CNT is an ideal experiment for studying the
effect of large electric fields on confinement.
Confinement is currently limited in CNT by the presence of
insulating rods in the plasma. The rods charge up electrostatically
and create an E x B convection of electrons out of the
magnetic surfaces. Confinement was studied as a function of
magnetic field, emitter bias, and neutral pressure. Electron loss
rate was found to be proportional to B^{-1} in the low neutral
pressure regime, which is consistent with rod-driven transport being
the dominant mechanism. At higher neutral pressures where the
neutral-driven transport is dominant, the loss rate scales linearly
with neutral pressure, as expected, and approximately as B^{-1.5}.
It is possible to use the measured data to separate the effects of
rod-driven and neutral-driven transport. Although the neutral-driven
confinement time is quite long (> 20 ms), it is still far less
than the theoretically predicted neoclassical confinement time. A
retractable electron emitter has been installed in CNT that creates
the plasma without having an insulating rod present in steady state.
Measurements of confinement time in this unperturbed plasma should
help to clearly identify the other transport mechanisms.
Characterization: A5
Comments:
