This paper documents work performed to convert scanned range data to CAD solid model representation. The work successfully developed surface fitting algorithms for quadric surfaces (e.g. plane, cone, cylinder, and sphere), and a segmentation algorithm based entirely on surface type, rather than on a differential metric like Gaussian curvature. Extraction of all CAD-required parameters for quadric surface representation was completed. Approximate face boundaries derived from the original point cloud were constructed. Work to extrapolate surfaces, compute exact edges and solid connectivity was begun, but left incomplete due to funding reductions. The surface fitting algorithms are robust in the face of noise and degenerate surface forms.
As part of a project with SEMATECH, detailed chemical reaction mechanisms have been developed that describe the gas-phase and surface chemistry occurring during the fluorocarbon plasma etching of silicon dioxide and related materials. The fluorocarbons examined are C{sub 2}F{sub 6}, CHF{sub 3} and C{sub 4}F{sub 8}, while the materials studied are silicon dioxide, silicon, photoresist, and silica-based low-k dielectrics. These systems were examined at different levels, ranging from in-depth treatment of C{sub 2}F{sub 6} plasma etch of oxide, to a fairly cursory examination of C{sub 4}F{sub 8} etch of the low-k dielectric. Simulations using these reaction mechanisms and AURORA, a zero-dimensional model, compare favorably with etch rates measured in three different experimental reactors, plus extensive diagnostic absolute density measurements of electron and negative ions, relative density measurements of CF, CF{sub 2}, SiF and SiF{sub 2} radicals, ion current densities, and mass spectrometric measurements of relative ion densities.
As computers become faster, have more memory, and use multiple parallel processors, large, complex codes that more accurately simulate physical phenomena have emerged to utilize this capability. Most problems can benefit from this approach and many require it. But not all! There are problems for which simpler methods on more modest computers still work. The trick is to identify those problems, write the codes, and make their implementation sufficiently simple that they can be used conveniently by those who could profit from them. A Simple Plasma Code has been written with this philosophy in mind. It retains just enough physics to allow realistic simulations to be formulated and run quickly, even on a personal computer. This paper describes the physical model, its numerical implementation, and presents a sample simulation.
Window and free surface interfaces perturb the flow in compression wave experiments. The velocity of these interfaces is routinely measured in shock-compression experiments using interferometry (i.e., VISAR). Interface perturbations often must be accounted for before meaningful material property results can be obtained. For shockless experiments when stress is a single valued function of strain, the governing equations of motion are hyperbolic and can be numerically integrated forward or backward in either time or space with assured stability. Using the VISAR results as ''initial conditions'' the flow fields are integrated backward in space to the interior of the specimen where the VISAR interface has not perturbed the flow at earlier times and results can be interpreted as if the interface had not been present. This provides a rather exact correction for free surface perturbations. The method can also be applied to window interfaces by selecting the appropriate initial conditions. Applications include interpreting Z-accelerator ramp wave experiments. The method applies to multiple layers and multiple reverberations. For an elastic-plastic material model the flow is dissipative and the governing equations are parabolic. When the parabolic terms are small, the equations also can be successfully integrated backward in space. This is verified by using a traditional elastic-plastic wave propagation code with a backward-derived stress history as the boundary condition for a forward calculation. Calculated free surface histories match the starting VISAR record verifying that the backward method produced an accurate solution to the governing equations. With our cooperation, workers at Los Alamos have successfully applied the Sandia-developed backward technique for the time-dependent quasielastic material model and are analyzing stress histories at a spall plane using the VISAR free surface velocity measurement from a ''pullback'' experiment.
Superresolution concepts offer the potential of resolution beyond the classical limit. This great promise has not generally been realized. In this study we investigate the potential application of superresolution concepts to synthetic aperture radar. The analytical basis for superresolution theory is discussed. The application of the concept to synthetic aperture radar is investigated as an operator inversion problem. Generally, the operator inversion problem is ill posed. A criterion for judging superresolution processing of an image is presented.
This report is a supplement to ''The Unique Signal Concept for Detonation Safety in Nuclear Weapons,'' SAND91-1269, which provides a prerequisite fundamental background on the unique signal (UQS) concept. The UQS is one of the key constituents of Enhanced Nuclear Detonation Safety (ENDS), as outlined in Section 1 of that report. There have been many documents written over the past quarter of a century describing various aspects of the UQS, but none of these emphasized the mathematical approaches that help explain why the UQS is effective in resisting inadvertent pre-arming, even in abnormal environments and how UQS implementations can be quantitatively assessed. The intent of this report is to describe various pertinent mathematical methodologies (many of which have not been previously reported) without duplicating, any more than necessary, background information available in other reports. Mathematical UQS analysis is needed because of quantitative requirements associated with ENDS, and because limited comparisons of various implementation approaches can be quantified under mathematical modeling assumptions. Some of the mathematics-based results shown in this report are presented to explain: (1) The reasons that the UQS methodology can provide greater protection against accident environments than could combinational techniques (Sections 2.1 through 2.4); (2) The reason that the probability of inadvertently duplicating a UQS comprising n bivalued events cannot be estimated as low as (1/2) inches (Section 2.4); (3) The value of, and the Sandia National Laboratories policy on independent sequential communication of UQS events (Section 3.4); and (4) The care that must be exercised if any signal processing is necessary (Section 4). There are also numerous examples (e.g., in Appendices A and B) of ill-advised deviations from UQS methodology that can seriously degrade safety. These examples help demonstrate that the UQS methodology should not be compromised.
Dual control volume molecular dynamics was employed to study the flux of methane through channels of thin silicalite membranes. The DCANIS force field was analyzed to describe the adsorption isotherms of methane and ethane in silicalite. The alkane parameters and silicalite parameters were determined by fiiting the DCANIS force field to single-component vapor-liquid coexistence curves (VLCC) and adsorption isotherms respectively. The adsorption layers on the surfaces of thin silicalite membranes showed a sifnificant resistance to the flux of methane. The results depicted the insensitivity of permeance to both the average pressure and pressure drop.
For highly cross-linked polymer networks bonded to a solid surface, the effect of interfacial bond density and system size on interfacial fracture is studied using molecular dynamics simulations. Results for tensile and shear mode simulations are given. The correspondence between the stress-strain curve and the sequence of molecular deformations is obtained. The failure strain for a fully bonded surface is equal to the strain necessary to make taut the average of the minimal paths through the network from a bonded site on the bottom solid surface to a bonded site on the top surface. At fractional interfacial bond densities, cavities form above the nonbonded surface, yielding an inhomogeneous strain profile and a smaller failure strain. The failure strain and stress are linearly proportional to the number of bonds at the interface except in the tensile mode when number of bonds is so few that van der Waals interactions dominate. The failure mode is successfully constructed to be interracial by limiting the interfacial bond density to be less than the bulk bond density.
Advanced thin-film processing and packaging technologies are employed in the fabrication of new planar thin-film multijunction thermal converters (MJTCs). The processing, packaging, and design features build on experience gained from prior NIST demonstrations of thin-film converters and are optimized for improved sensitivity, bandwidth, manufacturability, and reliability.