I have finished up a little RF ion source, at least to the point that it's operable. The idea behind such sources dates back to the late '40s and a physicist named Thonemann. An electrodeless high-frequency discharge is induced in a dielectric "bottle" and ions are extracted from this discharge by DC electric fields. The design I use is closely related to the Bayly design that was improved by Kiss and Koltay, whose paper I provided in a prior thread. The difference between this embodiment and Thonemann's is that both the DC extraction electrodes are located at the downstream side of the bottle. This avoids numerous practical difficulties, and for fusor injection, has a major upside: particles injected via a negative-HV "puller" electrode rather than by a positive-HV probe at the upstream end of the bottle will not have too much energy to remain trapped in the fusor.
So here's my take on the Kiss-Koltay RF ion source. The puller electrode has a 1.5 mm x 7 mm axial hole in it and is supported from a commercial low-profile insulator by a piece of copper tubing. The grounded electrode fits just outside the puller on another flange that also carries a "quick connect" compression fitting to seal against the glass bottle. An axial magnetic field in the discharge region helps confine plasma to the source axis where it is a useful supply of ions. RF power (~30 W) at 200 MHz is coupled into the discharge via an L-match network and two turns of wire.
Photo #1: Schematic of ion source
Photo #2: The ion source in operation in air; diffusion pump system off to left.
Photo #3: Assembly of source showing installation of the negative puller electrode on its carrier flange.
Photo #4: Addition of the grounded electrode over the top of puller.
Photo #5: The target for this test system is a graphite cylinder.
Photo #6: Close-up of beam, probably about 1-2 mA (I cannot yet measure this accurately). Pardon the annoying soap scum on the front of my viewport. The puller is at -3 kV, the target at about -5 kV. Electrons backstreaming from the target cause the low-profile insulator carrying the puller to fluoresce.
Photo #7: Another pic of the beam.
Unfortunately my little diffusion pump stand lacks the speed to cope with this differentially-pumped system per my design. The diff pump is restricted to about 5 lps by the KF25 plumbing in that system. Therefore, I have to run the ion source discharge at reduced pressure for this startup test. I am working on a Varian V-80 turbopump to carry on the trials, which should allow me to put deuterium in the capillary feed. Stay tuned!
-Carl
|
2008-14-09 04:09 RF ion source sees first light 
(Carl Willis) [Latest: 2009-09-01 17:18]