[en] The indirect-drive targets being considered for inertial fusion require the driver to deposit around 5 MJ on a target in less than 10 ns. This requirement can in principle be met by heavy-ion beams with particle masses between 120 and 240 amu, an ion kinetic energy in the range of 6--12 GeV, and a total current in excess of 30 kA. Three strategies for generating beams with these parameters are currently being studied. European laboratories are investigating the use of low-current beams from a radio-frequency accelerator. To obtain the needed current density, these beams would be stacked and accumulated in storage rings and then directed simultaneously at the target. American researchers are developing high-current induction accelerators, and the two principal configurations under consideration are the linear driver and the ''recirculator,'' in which ion pulses pass repeatedly through the same accelerator elements. The merits of the three approaches are compared, and key physics uncertainties in each are identified

[en] The indirect-drive targets being considered for inertial fusion require the driver to deposit 5 MJ on a target in less than 10 ns. This requirement can in principle be met by on beams with particle masses between 120 and 240 amu, an ion kinetic energy in the range of 6--12 GeV, and a total current in excess of 30 kA. Three strategies for generating beams with these parameters are currently being studied. European laboratories are investigating the use of low-current beams from a radio-frequency accelerator. To obtain the needed current density, these beams would be stacked and accumulated in storage rings and then directed simultaneously at the target. American researchers are developing high-current induction accelerators, and the two principal configurations under consideration are the linear driver and the ''recirculator,'' in which ion pulses pass repeatedly through the same accelerator elements. The merits of the three approaches are compared, and key physics uncertainties in each are identified

[en] A new analytic model is presented which accurately estimates the radially averaged axial component of the space-charge field of an axisymmetric heavy-ion beam in a cylindrical beam pipe. The model recovers details of the field near the beam ends that are overlooked by simpler models, and the results compare well to exact solutions of Poisson's equation. Field values are shown for several simple beam profiles and are compared with values obtained from simpler models

[en] Electrostatic bounce modes are standing waves that occur in a mirror plasma when the relative spread in electron bounce frequencies is small. The modes can be destabilized by an ion distribution with a peaked perpendicular energy, and experimental data suggest that this mechanism was the principal cause of instability in certain low-density mirror experiments. After a review of theoretical work on electrostatic waves in mirror plasmas, a general matrix eigenvalue equation for the wave potential is derived which accounts accurately for electron histories and which includes the ion response. A computer program for calculating the plasma eigenmodes and the associated threshold densities for instability and maximum growth rates is then described. The threshold densities for unstable bounce modes expected in the Baseball I and Baseball II devices are compared with experimental values. The good agreement between theoretical and experimental thresholds in Baseball II makes bounce modes the most likely cause of instabilities in that device. In Baseball I, the most unstable modes expected from the theory have threshold densities consistently below observed values. The discrepancy probably results from idealizations in the model that reduce wave damping

[en] Electrostatic bounce modes are standing waves that occur in a mirror plasma when the relative spread in electron bounce frequencies is small. The modes can be destabilized by an ion distribution with a peaked perpendicular energy, and experimental data suggest that this mechanism was the principal cause of instability in certain low-density mirror experiments. After a review of theoretical work on electrostatic waves in mirror plasmas, a general matrix eigenvalue equation for the wave potential is derived which accounts accurately for electron histories and which includes the ion response. A computer program for calculating the plasma eigenmodes and the associated threshold densities for instability and maximum growth rates is then described. The threshold densities for unstable bounce modes expected in the Baseball I and Baseball II devices are compared with experimental values. The good agreement between theoretical and experimental thresholds in Baseball II makes bounce modes the most likely cause of instabilities in that device. In Baseball I, the most unstable modes expected from the theory have threshold densities consistently below observed values

[en] Simulation of designs of an Integrated Beam Experiment (IBX) class accelerator have been carried out. These simulations are an important tool for validating such designs. Issues such as envelope mismatch and emittance growth can be examined in a self-consistent manner, including the details of injection, accelerator transitions, long-term transport, and longitudinal compression. The simulations are three-dimensional and time-dependent, and begin at the source. They continue up through the end of the acceleration region, at which point the data is passed on to a separate simulation of the drift compression. Results are be presented

[en] Recent advances in solid-state switches have made it feasible to design programmable, high-repetition-rate pulsers for induction accelerators. These switches could lower the cost of recirculating induction accelerators, such as the ''small recirculator'' at Lawrence Livermore National Laboratory (LLNL), by substantially reducing the number of induction modules. Numerical work is reported here to determine what effects the use of fewer pulsers at higher voltage would have on the beam quality of the LLNL small recirculator. Lattices with different numbers of pulsers are examined using the fluid/envelope code CIRCE, and several schedules for acceleration and compression are compared for each configuration. For selected schedules, the phase-space dynamics is also studied using the particle-in-cell code WARP3d

No abstract available

[en] An envelope code has been developed to facilitate the design of a recirculating accelerator for a heavy-ion fusion reactor. A novel feature of the model is the treatment of the beam charge density as a Lagrangian fluid in the axial direction. Transport results for a preliminary recirculator design are presented, and sensitivity of the transport to errors in the magnet strength is discussed. 4 refs., 4 figs

[en] Electron bounce modes can occur in mirror plasmas when the spread in the bounce frequency is small. This condition is satisfied in mirror devices when electrons are principally confined by an approximately quadratic electrostatic potential. These modes are examined by numerically solving an integral equation for the perturbed wave potential in a mirror plasma. A long wavelength mode is found that can be destabilized by ions because of the loss-cone nature of their distribution. Threshold densities and maximum growth rates are calculated using a perturbation method. The theoretical stability threshold predictions agree closely with Baseball II measurements