George D. Craig

Numerical simulation of intense-beam experiments at LLNL and LBNL

We present intense-beam simulations with the WARP code that are being carried out in support of the Heavy-Ion Fusion experimental programs at Lawrence Livermore National Laboratory (LLNL) and Lawrence Berkeley National Laboratory (LBNL). The WARP code is an electrostatic particle-in-cell code with an extensive hierarchy of simulation capabilities. Two experiments are analyzed. First, simulations are presented on an 80 keV, 2 mA K‘ bent transport channel at LLNL that employs an alternating-gradient lattice of magnetic quadrupoles for beam focusing and electric dipoles for beam bending. Issues on dispersion-induced changes in beam quality on the transition from straight- to bent-lattice sections are explored. The second experiment analyzed is a 2 MeV, 800 mA, driver-scale injector and matching section at LBNL that is based on a K‘ source and an alternating-gradient lattice of electrostatic quadrupoles biased to accelerate, focus, and match the beam. Issues on beam quality, space-charge waves, and beam hollowing are explored. Published by Elsevier Science B.V.

Progress Toward Source-to-Target Simulation

Source-to-target simulation of an accelerator provides a thorough check on the consistency of the design as well as a detailed understanding of the beam behavior. Issues such as envelope mis-match and emittance growth can be examined in a self-consistent manner, including the details of accelerator transitions, long-term transport, and longitudinal compression. The large range in scales, from centimeter-scale transverse beam size and applied field scale-length, to meter-scale beam length, to kilometer-scale accelerator length, poses a significant computational challenge. The ever-increasing computational power that is becoming available through massively parallel computers is making such simulation realizable. This paper discusses the progress toward source-to-target simulation using the WARP particle-in-cell code. Representative examples are shown, including 3-D, along-term transport simulations of Integrated Research Experiment (IRE) scale accelerators.

Results from the recirculator project at LLNL

The Heavy Ion Fusion Group at Lawrence Livermore National Laboratory has for several years been developing the worlds first circular induction accelerator designed for space charge dominated ion beams. Experiments on one quarter of the ring have been completed. The accelerator extended ten half-lattice periods (HLP) with induction cores for acceleration placed on every other HLP. A network of Capacitive Beam Probes (C-probes) was also enabled for beam position monitoring throughout the bend section. These C-probes have been instrumental in steering experiment, implementation of the acceleration stages and the dipole pulser, and the first attempts at coordinated bending and acceleration. Data from these experiments and emittance measurements will be presented.

Status of experiments leading to a small recirculator

A heavy ion linear induction accelerator is considered to be the leading driver candidate for an Inertial Fusion Energy reactor. To deliver a space-charge-dominated beam at the appropriate energy (several GeV), such an accelerator would be several kilometers in length. Since total length has a strong influence on accelerator cost, we are considering the potential advantages and practical implementation of a recirculating induction accelerator. To address the critical scientific and technical challenges of a recirculating space-charge-dominated heavy ion beam, we have begun to develop the elements of a scaled ``small recirculator``. An operating recirculator must demonstrate full beam control including multi-lap operation, beam insertion/extraction, acceleration and pulse compression. At present, experiments have been conducted using a 2mA, 80keV K{sup +} beam transported through a 45{degree} bend; experiments on a 90{degree} bend with five induction modulators will begin soon. This paper briefly summarizes the recirculator specifications and operational features and reports the latest experimental data as well as the developmental status of beam diagnostics.

Emittance growth in heavy ion rings due to the effects of space charge and dispersion

We review the derivation of moment equations which include the effects of space charge and dispersion in bends first presented in ref. [1]. These equations generalize the familiar envelope equations to include the dispersive effects of bends. We review the application of these equations to the calculation of the change in emittance resulting from a sharp transition from a straight section to a bend section, using an energy conservation constraint. Comparisons of detailed 2D and 3D simulations of intense beams in rings using the WARP code (refs. [2,3]) are made with results obtained from the moment equations. We also compare the analysis carried out in ref. [1], to more recent analyses, refs. [4,5]. We further examine self-consistent distributions of beams in bends and discuss the relevance of these distributions to the moment equation formulation.

On the re‐acceleration of bunched beams

We examine the re‐acceleration of a bunched beam through a linear induction accelerator (LIA) cavity, with attention to the energy lost through coupling to the TM modes of the structure. We find that the energy lost at 1 kA peak current is a small fraction of the boost which the LIA is designed to impart. We discuss implications for a Relativistic Klystron or Free Electron Laser (FEL) version of the Two‐Beam Accelerator (TBA).

Crystallization and X‐ray diffraction data for a new form of concanavalin A

A new crystal form of concanavalin A, a well studied lectin from the jack bean (Canavalia ensiformis) is reported. These crystals, with the symmetry of orthorhombic space group C222(1), grow as large roughly equi-dimensional blocks. Unit-cell parameters at 120 K are a = 118.67 (12), b = 101.36 (13), c = 111.94 (9) A. On density considerations for two molecules per asymmetric unit, the water content is approximately 60%. Data to a nominal resolution of 1.5 A were collected.