[en] Non-relativistic ring motion with constant acceleration occuring under the action of slightly nonhomogeneous magnetic or electric fields is considered. Ion losses take place as a result of field fluctuations. Field fluctuations lead to an appearance of an accidential inertial force in the ion oscillation equation. The oscillation amplitude second power may be increased up to its limiting value under the action of this force. Ion losses occur under action of electromagnetic field fluctuations. It is shown that the value of ion losses is determined by the formula derived for particle losses under action of a random force, in the framework of the acceleration theory

[en] A stability criterion for the tilt mode of a spheromak-ion ring hybrid configuration has been developed for the case where the ring current is small compared to the spheromak azimuthal current. It is shown that the stability is related to the distortion of the spheromak separatrix

No abstract available

[en] This thesis research used an annular, applied-B, extraction ion diode and cusped magnetic field (the Megavolt Ion Coil Experiment) designed to produce ion rings whose parameters approach those desired for tilt stabilization of other plasma confinement devices. The work consisted of three principal parts: (1) Design, construction, and testing of a magnetic cusp-field to convert a cylindrical beam into a hollow rotating beam. This magnet has worked perfectly for over 5 years. (2) Detailed numerical investigations of single-particle trajectories in such cusp fields. These simulations indicate that unless the beam is more than 99% space-charge neutralized (more than previously realized), the ion beam will not propagate through the cusp field. Analysis of beam/gas interactions show that under these conditions, the only significant gas ionization mechanism is direct impact ionization. (3) Experimental investigation of an ion beam's behavior in traversing a cusp. Unfortunately, due to funding limitations, these experiments had to be terminated prematurely, and only a limited data base could be acquired. Comparison of these data to simulations without self-fields indicated significant discrepancies: if the full ion current was carried by protons, the calculated downstream beam diamagnetic signals were an order of magnitude higher than the observed signals, and the time width was approximately half the observed width. The principal conclusions reaches are: (1) The ion energy appears to be 10% to 15% lower than the measured applied voltage, perhaps due to charge-exchange effects in the diode. (2) Space-charge fields of only a few kV/cm in the cusp can cause enough axial drag and particle loss to reproduce the observed low levels of downstream signals. (3) The longer duration of the observed signals may have been caused by plasma currents driven in the beam-generated plasma, although the data are not definitive

[en] Numerical studies of the injection and resistive trapping efficiency of ion rings, using an improved algorithm are presented. Trapping efficiency is found to be strongly dependent upon the number of particles injected and upon mirror ratios in the system. Wall resistivity and beam divergence affect the process to a lesser extent. (author)

[en] The Cornell ion ring program consists of experiments, numerical simulation, and theory designed to show the way toward, and eventually achieve, the goal of providing ion rings for creating and/or maintaining, stabilizing and heating compact toroidal confinement devices. Summarized are the experimental results to date and briefly mentioned are simulation results directly relevant to the experiments

No abstract available

[en] A detailed study of the trapping of proton rings in a magnetic mirror both in uniform gas fillings and in localized puffed-gas fillings is presented. The trapping of about 3 x 10¹⁴, 400 keV protons in a 1.5 m long, 0.8 T magnetic mirror was accomplished with a static 1.23:1 downstream mirror and a fast ''gate'' upstream mirror with mirror ratio <1.3:1. Protons were trapped for 4 μs (>50 cyclotron periods) or longer. The proton ring mean radius, length, and annular thickness in the trap were 10 cm, 1.5 m, and 5--8 cm, respectively. Detailed study of the dynamics of the injected and trapped protons with a variety of diagnostic techniques showed that the bulk of the protons followed trajectories that could be fully explained using the ion diode voltage and the details of the applied magnetic field configuration. However, a small fraction of the injected protons had axial energy much higher than was possible from single-particle trajectories; incomplete electrostatic neutralization of the proton ring by the beam-generated plasma in the injection region is postulated to explain this observation. Ring proton lifetime in the mirror trap appeared to be determined by mirror losses upon initially encountering the downstream and then the upstream mirrors, followed by a gradual radial loss due to charge-exchange collisions with the neutral gas contained in the experiment chamber

[en] Functional principles, structure, and the parameters of the prototype, some trends of development, and possible applications of the heavy-ion collective accelerator are described. (author)

[en] A new, 3-D electromagnetic (EM), hybrid, particle-in-cell (PIC) code, FLAME has been constructed to study low-frequency, large orbit plasmas in realistic cylindrical configurations. The stability and equilibrium of strong ion rings in magnetized plasmas are the first issues suitable for its application. In FLAME the EM-field is governed by Maxwell's equations in the quasi-neutral Darwin approximation (with displacement current neglected), the ion components are represented by discrete macro-particles, and the plasma electrons are modeled as a massless cold fluid. All physical quantities are expanded into finite Fourier series in the azimuthal (θ) direction. The discretization in the poloidal (r, z) plane is done by a finite-difference staggered grid method. The electron fluid equations include a finite scalar resistivity and macro-particles experience slowing-down collisions. A substantial reduction of computation time is achieved by enabling separate time advances of background and beam particle species in the time-averaged fields. FLAME has been optimized to run on parallel, MIMD systems, and has an object-oriented (C++) structure. The results of normal mode tests intended to verify the code ability to correctly model plasma phenomena are presented. The authors also investigate in 3-D the injection of a powerful annular ion beam into a plasma immersed in a magnetic cusp followed by an axially ramped applied magnetic field. A nonaxisymmetric perturbation is applied to the magnetic field and its effect on ion ring formation is analyzed. 28 refs., 13 figs., 1 tab