The idea of producing beams of high energy photons by Compton backscattering of laser photons was proposed over 20 years ago. At the time, producing the required laser pulses was not feasible. However, recent advances in high average power, diode pumped lasers appear to have solved this problem. The US Collaboration is now turning its attention to the engineering requirement of mating the laser and optics components with the accelerator structures in the confined space of the a colliding beam interaction region. The demonstration of a technically feasible interaction region design is planned for the Snowmass conference in 2001.

The Plutonium Immobilization Program requires development of the process and plant prototypic equipment to immobilize surplus plutonium in ceramic for long-term storage. Because of the hazardous nature of plutonium, it was necessary to develop a remotely operable materials transfer system which can function within the confines of a glovebox. In support of this work at LLNL, such a material transfer system (MTS) was developed. This paper presents both the mechanical and controls parts making up this system, and includes photographs of the key components and diagrams of their assemblies, as well as a description of the control sequence used to validate the MTS capabilities.

This paper discusses: (1) A general block-factorization (matrix) form of multilevel preconditioners; algebraic methods; (2) Selecting parameters based on the matrix topology; graph based algorithms; (3) Examples of coarsening; (4) Numerical experiments.

ParaDyn is a parallel version of the DYNA3D computer program, a three-dimensional explicit finite-element program for analyzing the dynamic response of solids and structures. The ParaDyn program has been used as a production tool for over three years for analyzing problems which range in size from a few tens of thousands of elements to between one-million and ten-million elements. ParaDyn runs on parallel computers provided by the Department of Energy Accelerated Strategic Computing Initiative (ASCI) and the Department of Defense High Performance Computing and Modernization Program. Preprocessing and post-processing software utilities and tools are designed to facilitate the generation of partitioned domains for processors on a massively parallel computer and the visualization of both resultant data and boundary data generated in a parallel simulation. This manual provides a brief overview of the parallel implementation; describes techniques for running the ParaDyn program, tools and utilities; and provides examples of parallel simulations.

A technique is developed to construct bulk hydroxyapatite (HAp) with different cellular structures. The technique involves the initial synthesis of nanocrystalline hydroxyapatite powder from an aqueous solution using water-soluble compounds and then followed by spray drying into agglomerated granules. The granules were further cold pressed and sintered into bulks at elevated temperatures. The sintering behavior of the HAp granules was characterized and compared with those previously reported. Resulting from the fact that the starting HAp powders were extremely fine, a relatively low activation energy for sintering was obtained. In the present study, both porous and dense structures were produced by varying powder morphology and sintering parameters. Porous structures consisting of open cells were constructed. Sintered structures were characterized using scanning electron microscopy and x-ray tomography. In the present paper, hydroxyapatite coatings produced by magnetron sputtering on silicon and titanium substrates will also be presented. The mechanical properties of the coatings were measured using nanoindentation techniques and microstructures examined using transmission electron microscopy.

Use of multiple backlighter foils and/or double-sided laser interaction geometry with backlit imaging can result in improved backlighter efficiency. An experimental comparison of backlighter intensity for Ti foils and Ti/Sc combination foils in both the one-sided and double-sided laser-interaction configuration is presented. Spectrally-integrated framing camera images show intensity contributions of front and rear backlighter surfaces for both foil types. Analysis of time-resolved x-ray spectra collected from foil targets show the relative contribution of Ti and Sc 2-1 He-like resonance lines to the total backlighter intensity.

{l_brace}311{r_brace} defects and dislocation loops are formed after ion-implantation and annealing of a silicon wafer. Recent Transmission Electron Microscopy studies by Li and Jones have shown that sub-threshold dislocation loops nucleate from {l_brace}311{r_brace} defects. In our study, the conjugate gradient method with the Stillinger Weber potential is used to relax different configurations such as {l_brace}311{r_brace} defects with a maximum of five chains and perfect dislocation loops. From the formation energies thus obtained we find that there is an optimal width for each length of the {l_brace}311{r_brace} defects. Moreover the relative stability of {l_brace}311{r_brace}s and loops is studied as a function of defect size. We observe that at very small sizes the perfect loops are more stable than the {l_brace}311{r_brace}s. This may provide an explanation for the experimental observation by Robertson et al that, in an annealing study of end of range damage of amorphized samples, 45% of the loops had nucleated in the first 10 minutes of anneal. We propose that homogeneous nucleation, as against unfaulting of the {l_brace}311{r_brace}s, could be the source of these loops.

We are attempting to characterize and model the micromechanical response of highly-filled polymers. In this class of materials, the continuous plastic binder used to bond the highly-filled material dominates the observed viscoelastic response. As a result, realistic lifetime analysis of these materials will require a thorough understanding of the contribution of the plastic binder. Laboratory applications of these materials include plastic bonded explosives, propellants, a variety of specialized filled organic materials for stockpile systems, and highly filled epoxy dielectric materials for the National Ignition Facility. We have explored numerous techniques to characterize the local microstructure of plastic bonded explosives. However, insufficient funding was obtained to bring these technologies to maturity, nevertheless our present tool set is significantly better than 2 years ago. We have also made some progress in developing an appropriate micromechanical constitutive modeling framework, based on a finite element method incorporating a cohesive zone model to represent the binder contribution within a Voronoi tesselation mesh structure for the PBX grains. A second modeling approach was used to incorporate analytical micromechanics (generalized self-consistent schemes). However, preliminary theoretical analysis strongly suggested that this approach would be invalid for such extremely high-filled systems like PBX.

Large-scale scientific computation, and all of the disciplines that support it and help to validate it, have been placed at the focus of Lawrence Livermore National Laboratory by the Accelerated Strategic Computing Initiative (ASCI). The Laboratory operates the computer with the highest peak performance in the world and has undertaken some of the largest and most compute-intensive simulations ever performed. Computers at the architectural extremes, however, are notoriously difficult to use efficiently. Even such successes as the Laboratory's two Bell Prizes awarded in November 1999 only emphasize the need for much better ways of interacting with the results of large-scale simulations. Advances in scientific computing research have, therefore, never been more vital to the core missions of the Laboratory than at present. Computational science is evolving so rapidly along every one of its research fronts that to remain on the leading edge, the Laboratory must engage researchers at many academic centers of excellence. In FY 1999, the Institute for Scientific Computing Research (ISCR) has expanded the Laboratory's bridge to the academic community in the form of collaborative subcontracts, visiting faculty, student internships, a workshop, and a very active seminar series. ISCR research participants are integrated almost seamlessly with the Laboratory's …

A wireless link was designed using a patch antenna. In the process, several different models were tested. Testing proved a patch antenna was a viable solution for building a wireless link within the design specifications. Also, this experimentation provided a basis for future patch antenna design.

The nonlinear Poisson-Boltzmann (PB) equation is solved using Pseudo Transient Continuation. The PB solver is constructed by modifying the nonlinear diffusion module of a 3D, massively parallel, unstructured-grid, finite element, radiation-hydrodynamics code. The solver also computes the electrostatic energy and evaluates the force on a user-specified contour. Either Dirichlet or mixed boundary conditions are allowed. The latter specifies surface charges, approximates far-field conditions, or linearizes conditions ''regulating'' the surface charge. The code may be run in either Cartesian, cylindrical, or spherical coordinates. The potential and force due to a conical probe interacting with a flat plate is computed and the result compared with direct force measurements by chemical force microscopy.

The interior voltage of a large metal can with thick walls struck directly by lightning was estimated using diffusion theory, aperture slot voltage theory, and experimental data. The hollow cylinder is closed at both ends. One end has a cap that is welded to the cylinder wall making a continuous electrical interface. The other end consists of a circular plate that is pressed into the cylinder wall and held under pressure with a threaded ring. From our experience with coupling measurements, this joint will be a weak link. It will allow more current to leak into the interior than from diffusion through the walls. Because the joint was designed for mechanical purposes, the electrical properties, such as continuity around the circumference, are not well controlled. Therefore, it is difficult to determine a single voltage attributed to this joint design with varying electrical characteristics. Instead, we will make a best effort of bounding the problem using both analytical calculations and data from tests of similar structures. The calculated internal cylinder voltage subjected to an extreme lightning strike from current diffusing through the wall is 19 volts. We estimate that the press-fit end plate will increase this voltage by a factor of …

This work investigates a control system for HCCI engines, where thermal energy from exhaust gas recirculation (EGR) and compression work in the supercharger are either recycled or rejected as needed. HCCI engine operation is analyzed with a detailed chemical kinetics code, HCT (Hydrodynamics, Chemistry and Transport), that has been extensively modified for application to engines. HCT is linked to an optimizer that determines the operating conditions that result in maximum brake thermal efficiency, while meeting the restrictions of low NO{sub x} and peak cylinder pressure. The results show the values of the operating conditions that yield optimum efficiency as a function of torque and RPM. For zero torque (idle), the optimizer determines operating conditions that result in minimum fuel consumption. The optimizer is also used for determining the maximum torque that can be obtained within the operating restrictions of NO{sub x} and peak cylinder pressure. The results show that a thermally controlled HCCI engine can successfully operate over a wide range of conditions at high efficiency and low emissions.

During the WBS4 review almost a year ago in January, and in more recent discussions of needed capsule characterization, I was struck by three pieces of information related to machined Be capsules: (1) that the speed of sound (and perhaps other properties) in Be grains is markedly different in orthogonal directions, (2) that the grain size in the best refined bulk Be samples is 10 to 15 {micro}m, and that these grains in the best case are randomly oriented, and (3) that we are concerned about capsule homogeneity on the level of 1 part in 10{sup 4}, presumably over angular length scales corresponding to the maximum in the growth curve. It seems to me that the first two points might be inconsistent with the third, and this led me to attempt to model the effect of randomly oriented Be grains on the radially dependent properties of a capsule.

The Inner and Outer modules of the Central Solenoid Model Coil (CSMC) were built by US and Japanese home teams in collaboration with European and Russian teams to demonstrate the feasibility of a superconducting Central Solenoid for ITER and other large tokamak reactors. The CSMC mass is about 120 t, OD is about 3.6 m and the stored energy is 640 MJ at 46 kA and peak field of 13 T. Testing of the CSMC and the CS Insert took place at Japan Atomic Energy Research Institute (JAERI) from mid March until mid August 2000. This paper presents the main results of the tests performed.

We present results from the Sustained Spheromak Physics Experiment (SSPX) at LLNL, which has been built to study energy confinement in spheromak plasmas sustained for up to 2 ms by coaxial DC helicity injection. Peak toroidal currents as high as 600kA have been obtained in the 1m dia. (0.23m minor radius) device using injection currents between 200-400kA; these currents generate edge poloidal fields in the range of 0.2-0.4T. The internal field and current profiles are inferred from edge field measurements using the CORSICA code. Density and impurity control is obtained using baking, glow discharge cleansing, and titanium gettering, after which long plasma decay times ({tau} {ge} 1.5ms) are observed and impurity radiation losses are reduced from {approx}50% to <20% of the input energy. Thomson scattering measurements show peaked electron temperature and pressure profiles with T{sub e} (0){approx}120eV and {beta}{sub e}{approx}7%. Edge field measurements show the presence of n=1 modes during the formation phase, as has been observed in other spheromaks. This mode dies away during sustainment and decay so that edge fluctuation levels as low as 1% have been measured. These results are compared with numerical simulations using the NIMROD code.

We present a new approach for investigating the kinetics of defect migration during annealing of low-energy, ion-implanted silicon, employing a combination of computer simulations and atomic-resolution tunneling microscopy. Using atomically-clean Si(111)-7x7 as a sink for bulk point defects created by 5 keV Xe and Ar irradiation, we observe distinct, temperature-dependent surface arrival rates for vacancies and interstitials. A combination of simulation tools provides a detailed description of the processes that underly the observed temperature-dependence of defect segregation, and the predictions of the simulations agree closely with the experimental observations.

The intense ion beams in a heavy ion Inertial Fusion Energy (IFE) driver and fusion chamber are non-neutral plasmas whose dynamics are largely dominated by space charge. We propose to develop a ''source-to-target'' Heavy Ion Fusion (HIF) beam simulation capability: a description of the kinetic behavior of this complex, nonlinear system which is both integrated and detailed. We will apply this new capability to further our understanding of key scientific issues in the physics of ion beams for IFE. The simulations will entail self-consistent field descriptions that require interprocessor communication, but are scalable and will run efficiently on terascale architectures. This new capability will be based on the integration of three types of simulations, each requiring terascale computing: (1) simulations of acceleration and confinement of the space-charge-dominated ion beams through the driver (accelerator, pulse compression line, and final focusing system) which accurately describe their dynamics, including emittance growth (phase-space dilution) effects; these are particle-in-cell (PIC) models; (2) electromagnetic (EM) and magnetoinductive (Darwin) simulations which describe the beam and the fusion chamber environment, including multibeam, neutralization, stripping, beam and plasma ionization processes, and return current effects; and (3) highly detailed simulations (6f, multispecies PIC, continuum Vlasov), which can examine electron effects …

Linear stability analysis of a two-cylinder approximation to gun injection--one cylinder to represent the confined spheromak and another to represent the gun--is shown to yield equilibria in which tearing modes exist simultaneously at the magnetic axis and at the geometric (gun) axis, as might be required to sustain helicity injection. These equilibria are MHD stable at the two axes but may have localized MHD instability at an interior minimum in the q profile. The theory predicts two tearing thresholds with successively deeper q minima as the gun current is increased at constant bias flux.

None

Initial measurements are reported for the Mercury laser system, a scalable driver for rep-rated high energy density physics research. The performance goals include 10% electrical efficiency at 10 Hz and 100 J with a 2-10 ns pulse length.

The utilization of high-power short pulse laser employing chirped-pulse amplification (CPA) for material processing and inertial confinement research is widely increasing. The performance of these high-power CPA laser system continues to be limited by the ability of the pulse compression gratings to hold up to the high-average-power or high-peak-power of the laser. Pulse compression gratings used in transmission and fabricated out of bulk fused silica have intrinsically the highest laser damage threshold when compared with metal or multilayer dielectric gratings that work in reflection. LLNL has developed processing capability to produce high efficiency fused silica transmission gratings at sizes useful to future Petawatt-class systems, and has demonstrated high efficiency at smaller aperture. This report shows that fused silica diffraction exhibiting >95% efficiency into the -1 diffraction order in transmission (90{sup o} deflection of the incident light, at an incidence angle of 45{sup o} to the grating face). The microstructure of this grating consisted of grooves ion-beam etched to a depth of 1.6 microns with a pitch of 0.75 microns, using a holographically produced photoresist mask that was subsequently stripped away in significance to the fabrication of the small scale high efficiency grating was the development of the processing technology and …

We present a method for accelerating time dependent Monte Carlo radiative transfer calculations by using a discretization of the diffusion equation to calculate probabilities that are used to advance particles in regions with small mean free path. The method is demonstrated on problems with on 1 and 2 dimensional orthogonal grids. It results in decreases in run time of more than an order of magnitude on these problems, while producing answers with accuracy comparable to pure IMC simulations. We call the method Implicit Monte Carlo Diffusion, which we abbreviate IMD.

Recent work in heavy-ion fusion accelerators and final focusing systems shows a trend towards less current per beam, and thus, a greater number of beams. Final focusing magnets are susceptible to nuclear heating, radiation damage, and neutron activation. The trend towards more beams, however, means that there can be less shielding for each magnet, Excessive levels of nuclear heating may lead to magnet quench or an intolerable recirculating power for magnet cooling. High levels of radiation damage may result in short magnet lifetimes and low reliability. Finally, neutron activation of the magnet components may lead to difficulties in maintenance, recycling, and waste disposal. The present work expands upon previous, three-dimensional magnet shielding calculations for a modified version of the HYLIFE-I1 IFE power plant design. We present key magnet results as a function of the number of beams.