Ion trajectories in a coaxial ion diode with magnetic confinement, and an axial anode surrounded by a magnetically insulated cylindrical cathode

Summary form only given. Project TedibeAr is a two-phase scientific demonstration of means to trap and magnetically accelerate high density ion rings. It was conceived by Pharis Williams. TedibeAr-I is to demonstrate pulses of 10/sup 15/ ions at 300 keV energies. TedibeAr-II is to demonstrate pulses of 10/sup 18/ ions at 300 keV energies. Initial experiments at Texas A&M are using a planar ion diode with a flash-over anode and magnetically insulated virtual cathode to inject ions axially through a magnetic cusp for trapping in a cylindrical region having a uniform axial magnetic field. In this geometry, ions follow a roughly helical axis-encircling path between magnetic mirrors and may be accelerated by dynamically increasing the magnetic field. Energies of 5 MeV are expected with only partial field reversal. To achieve TedibeAr-II goals, flux reversal will be necessary. The coaxial ion diode recently designed by Greenly at Cornell University shows promise. Its most desirable feature is that ions are born trapped after radial injection through the cylindrical virtual cathode. In a uniform axial magnetic field at low ion density, 300 keV ions follow hypocycloidal non-axis-encircling orbits. Ions start near a 5-cm diameter anode and are accelerated radially through a 6-cm diameter virtual cathode. Each orbit crosses the cathode twice during its orbital period of /spl sim/50 to 100 nsec. To make betatron acceleration feasible, the ions need to follow predictable trajectories for hundreds of orbits. This is unlikely with uniform, time-invariant electric and magnetic fields. The self-fields of the intense ion beam will create spatial and temporal gradients. Some spatial and temporal gradients in the electric and magnetic fields tend toward stability and others tend toward instability of desired orbits. The design challenge is to blend the external and induced fields to produce the desired acceleration. This paper discusses some conditions for "ring" stability.