The gridded electromagnet probe

Abstract

We attempted to measure the anisotropy in the electron distribution function in magnetized plasma by exploiting the adiabatic invariance of the electron's magnetic moment with a probe comprising a grid, a collector, and an inertially cooled electromagnet. The electric mirror force of the grid, which is located at the electromagnet throat, reduces the demand on the magnetic mirror force and thereby on the electromagnet current, which then allows for the construction of a compact probe that can be inserted inside the plasma chamber. An analysis of the effects of space charge inside the grid-collector cavity revealed that the size of the probe's entrance aperture, b, which gives the size of the plasma beam inside the probe, should be chosen to be within a factor of ten of the electron Debye-length [lambda][sub]De. In addition, an analysis of the discrete structure of the grid showed that the mesh wire spacing d should be chosen to be much less than [lambda][sub]De. Also, the wire thickness t should be chosen to be much less than d . We built a probe with a grid of tungsten wires with dimensions, t = 5,[mu]m and d = 200[mu]m . We then tested this probe in a hydrogen plasma immersed in a background magnetic field of B [approximately] 1kG. The plasma was heated by microwaves via the electron cyclotron resonance. It was characterized by a density and temperature equal to n[sub]e [approximately] 10¹⁰ cm⁻³ and T[sub]e [approximately] 10eV, respectively, which gave [lambda][sub]De [approximately] 300[mu]m. The collector's current-voltage characteristic demonstrated the interaction between the electric barrier at the collector and the hybrid electric-magnetic barrier at the grid, thereby establishing the basic principles of the probe.

(cont.) The characteristic also revealed the non-ideal behaviors associated with the electric hole in the mesh and the effects of space charge. These effects in conjunction with the poor signal-to-noise level of the data prevented the measurement of the distribution function. Still, we were able to extract the temperature anisotropy for an assumed two-temperature Maxwellian distribution. The value for this ratio was found to be greater than one (greater temperature for the perpendicular gyro-motion), which is plausible given the way in which the plasma is heated.