Publications

On March 20, 1998, Sandia National Laboratories performed a double-blind test of the DKL LifeGuard human presence detector and tracker. The test was designed to allow the device to search for individuals well within the product`s published operational parameters. The Test Operator of the DKL LifeGuard was provided by the manufacturer and was a high-ranking member of DKL management. The test was developed and implemented to verify the performance of the device as specified by the manufacturer. The device failed to meet its published specifications and it performed no better than random chance.

Despite the best preventative measures, ruptured hoses, spills and leaks occur with use of all hydraulic equipment. Although these releases do not usually produce a RCRA regulated waste, they are often a reportable occurrence. Clean-up and subsequent administrative procedure involves additional costs, labor and work delays. Concerns over these releases, especially related to Sandia National Laboratories (SNL) vehicles hauling waste on public roads prompted Fleet Services (FS) to seek an alternative to the standard petroleum based hydraulic fluid. Since 1996 SNL has participated in a pilot program with the University of Iowa (UNI) and selected vehicle manufacturers, notably John Deere, to field test hydraulic fluid produced from soybean oil in twenty of its vehicles. The vehicles included loaders, graders, sweepers, forklifts and garbage trucks. Research was conducted for several years at UNI to modify and market soybean oils for industrial uses. Soybean oil ranks first in worldwide production of vegetable oils (29%), and represents a tremendous renewable resource. Initial tests with soybean oil showed excellent lubrication and wear protection properties. Lack of oxidative stability and polymerization of the oil were concerns. These concerns were being addressed through genetic alteration, chemical modification and use of various additives, and the improved lubricant is in the field testing stage.

The aim of this laboratory-directed research and development project was to study amorphous carbon (a-C) thin films for eventual cold-cathode electron emitter applications. The development of robust, cold-cathode emitters are likely to have significant implications for modern technology and possibly launch a new industry: vacuum micro-electronics (VME). The potential impact of VME on Sandia`s National Security missions, such as defense against military threats and economic challenges, is profound. VME enables new microsensors and intrinsically radiation-hard electronics compatible with MOSFET and IMEM technologies. Furthermore, VME is expected to result in a breakthrough technology for the development of high-visibility, low-power flat-panel displays. This work covers four important research areas. First, the authors studied the nature of the C-C bonding structures within these a-C thin films. Second, they determined the changes in the film structures resulting from thermal annealing to simulate the effects of device processing on a-C properties. Third, they performed detailed electrical transport measurements as a function of annealing temperature to correlate changes in transport properties with structural changes and to propose a model for transport in these a-C materials with implications on the nature of electron emission. Finally, they used scanning atom probes to determine important aspects on the nature of emission in a-C.

Net erosion rates of carbon target plates have been measured in situ for the DIII-D lower divertor. The principal method of obtaining this data is the DiMES sample probe. Recent experiments have focused on erosion at the outer strike-point of two divertor plasma conditions: (1) attached (Te > 40 eV) ELMing plasmas and (2) detached (Te < 2 eV) ELMing plasmas. The erosion rates for the attached cases are > 10 cm/year, even with incident heat flux < 1 MW/m{sup 2}. In this case, measurements and modeling agree for both gross and net carbon erosion, showing the near-surface transport and redeposition of the carbon is well understood and that effective sputtering yields are > 10%. In ELM-free discharges, this erosion rate can account for the rate of carbon accumulation in the core plasma. Divertor plasma detachment eliminates physical sputtering, while spectroscopically measured chemical erosion yields are also found to be low (Y(C/D{sup +}) {le} 2.0 {times} 10{sup {minus}3}). This leads to suppression of net erosion at the outer strike-point, which becomes a region of net redeposition ({approximately} 4 cm/year). The private flux wall is measured to be a region of net redeposition with dense, high neutral pressure, attached divertor plasmas. Leading edges intercepting parallel heat flux ({approximately} 50 MW/m{sup 2}) have very high net erosion rates ({approximately} 10 {micro}m/s) at the OSP of an attached plasma. Leading edge erosion, and subsequent carbon redeposition, caused by tile gaps can account for half of the deuterium codeposition in the DIII-D divertor.

This paper describes the design and design issues associated with silicon surface micromachined device design Some of the tools described are adaptations of macro analysis tools. Design issues in the microdomain differ greatly from design issues encountered in the macrodomain. Microdomain forces caused by electrostatic attraction, surface tension, Van der Walls forces, and others can be more significant than inertia, friction, or gravity. Design and analysis tools developed for macrodomain devices are inadequate in most cases for microdomain devices. Microdomain specific design and analysis tools are being developed, but are still immature and lack adequate functionality. The fundamental design process for surface micromachined devices is significantly different than the design process employed in the design of macro-sized devices. In this paper, MEMS design will be discussed as well as the tools used to develop the designs and the issues relating fabrication processes to design. Design and analysis of MEMS devices is directly coupled to the silicon micromachining processes used to fabricate the devices. These processes introduce significant design limitations and must be well understood before designs can be successfully developed. In addition, some silicon micromachining fabrication processes facilitate the integration of silicon micromachines with microelectronics on-chip. For devices requiring on-chip electronics, the fabrication processes introduce additional design constraints that must be taken into account during design and analysis.

Many manufacturing companies today expend more effort on upgrade and disposal projects than on clean-slate design, and this trend is expected to become more prevalent in coming years. However, commercial CAD tools are better suited to initial product design than to the product`s full life cycle. Computer-aided analysis, optimization, and visualization of life cycle assembly processes based on the product CAD data can help ensure accuracy and reduce effort expended in planning these processes for existing products, as well as provide design-for-lifecycle analysis for new designs. To be effective, computer aided assembly planning systems must allow users to express the plan selection criteria that apply to their companies and products as well as to the life cycles of their products. Designing products for easy assembly and disassembly during its entire life cycle for purposes including service, field repair, upgrade, and disposal is a process that involves many disciplines. In addition, finding the best solution often involves considering the design as a whole and by considering its intended life cycle. Different goals and constraints (compared to initial assembly) require one to re-visit the significant fundamental assumptions and methods that underlie current assembly planning techniques. Previous work in this area has been limited to either academic studies of issues in assembly planning or applied studies of life cycle assembly processes, which give no attention to automatic planning. It is believed that merging these two areas will result in a much greater ability to design for; optimize, and analyze life cycle assembly processes.

An agile microsystem manufacturing technology has been developed that provides unprecedented 5 levels of independent polysilicon surface-micromachine films for the designer. Typical surface-micromachining processes offer a maximum of 3 levels, making this the most complex surface-micromachining process technology developed to date. Leveraged from the extensive infrastructure present in the microelectronics industry, the manufacturing method of polysilicon surface-micromachining offers similar advantages of high-volume, high-reliability, and batch-fabrication to microelectromechanical systems (MEMS) as has been accomplished with integrated circuits (ICs). These systems, comprised of microscopic-sized mechanical elements, are laying the foundation for a rapidly expanding, multi-billion dollar industry 2 which impacts the automotive, consumer product, and medical industries to name only a few.

Composite doublers, or repair patches, provide an innovative repair technique which can enhance the way aircraft are maintained. Instead of riveting multiple steel or aluminum plates to facilitate an aircraft repair, it is possible to bond a single boron-epoxy composite doubler to the damaged structure. In order for the use of composite doublers to achieve widespread use in the civil aviation industry, it is imperative that methods be developed which can quickly and reliably assess the integrity of the doubler. In this study, a specific composite application was chosen on an L-1011 aircraft in order to focus the tasks on application and operation issues. Primary among inspection requirements for these doublers is the identification of disbonds, between the composite laminate and aluminum parent material, and delaminations in the composite laminate. Surveillance of cracks or corrosion in the parent aluminum material beneath the doubler is also a concern. No single nondestructive inspection (NDI) method can inspect for every flaw type, therefore it is important to be aware of available NDI techniques and to properly address their capabilities and limitations. A series of NDI tests were conducted on laboratory test structures and on full-scale aircraft fuselage sections. Specific challenges, unique to bonded composite doubler applications, were highlighted. An array of conventional and advanced NDI techniques were evaluated. Flaw detection sensitivity studies were conducted on applicable eddy current, ultrasonic, X-ray and thermography based devices. The application of these NDI techniques to composite doublers and the results from test specimens, which were loaded to provide a changing flaw profile, are presented in this report. It was found that a team of these techniques can identify flaws in composite doubler installations well before they reach critical size.

This report describes ship accident event trees, ship collision and ship fire frequencies, representative ships and shipping practices, a model of ship penetration depths during ship collisions, a ship fire spread model, cask to environment release fractions during ship collisions and fires, and illustrative consequence calculations. This report contains the following appendices: Appendix 1 -- Representative Ships and Shipping Practices; Appendix 2 -- Input Data for Minorsky Calculations; Appendix 3 -- Port Ship Speed Distribution; and Appendix 4 -- Cask-to-Environment Release Fractions.

The engineering of advanced semiconductor heterostructure materials and devices requires a detailed understanding of, and control over, the structure and properties of semiconductor materials and devices at the atomic to nanometer scale. Cross-sectional scanning tunneling microscopy has emerged as a unique and powerful method to characterize structural morphology and electronic properties in semiconductor epitaxial layers and device structures at these length scales. The basic experimental techniques in cross-sectional scanning tunneling microscopy are described, and some representative applications to semiconductor heterostructure characterization drawn from recent investigations in the authors laboratory are discussed. Specifically, they describe some recent studies of InP/InAsP and InAsP/InAsSb heterostructures in which nanoscale compositional clustering has been observed and analyzed.

On July 1--2, 1997, Sandia National Laboratories hosted the External Committee to Evaluate Sandia`s Risk Expertise. Under the auspices of SIISRS (Sandia`s International Institute for Systematic Risk Studies), Sandia assembled a blue-ribbon panel of experts in the field of risk management to assess their risk programs labs-wide. Panelists were chosen not only for their own expertise, but also for their ability to add balance to the panel as a whole. Presentations were made to the committee on the risk activities at Sandia. In addition, a tour of Sandia`s research and development programs in support of the US Nuclear Regulatory Commission was arranged. The panel attended a poster session featuring eight presentations and demonstrations for selected projects. Overviews and viewgraphs from the presentations are included in Volume 1 of this report. Presentations are related to weapons, nuclear power plants, transportation systems, architectural surety, environmental programs, and information systems.

An exploratory study was conducted under the Architectural Surety Program to examine the possibility of modifying fracture of glass in the shock-wave environment associated with terrorist bombings. The intent was to explore strategies to reduce the number and severity of injuries resulting from those attacks. The study consisted of a series of three experiments at the Energetic Materials Research and Testing Center (EMRTC) of the New Mexico Institute of Mining and Technology at Socorro, NM, in which annealed and tempered glass sheets were exposed to blast waves at several different levels of overpressure and specific impulse. A preliminary assessment of the response of tempered glass to the blast environment suggested that inducing early failure would result in lowering fragment velocity as well as reducing the loading from the window to the structure. To test that possibility, two different and novel procedures (indentation flaws and spot annealing) were used to reduce the failure strength of the tempered glass while maintaining its ability to fracture into small cube-shaped fragments. Each experiment involved a comparison of the performance of four sheets of glass with different treatments.

SPH (Smoothed Particle Hydrodynamics) is a gridless Lagrangian technique which is appealing as a possible alternative to numerical techniques currently used to analyze high deformation impulsive loading events. Previously, the SPH algorithm has been subjected to detailed testing and analysis to determine the feasibility of using the coupled finite-element/SPH code PRONTO/SPH for the analysis of various types of underwater explosion problems involving fluid-structure and shock-structure interactions. Here, SPH and Eulerian simulations are used to study the details of underwater bubble collapse, particularly the formation of re-entrant jets during collapse, and the loads generated on nearby structures by the jet and the complete collapse of the bubble. Jet formation is shown to be due simply to the asymmetry caused by nearby structures which disrupt the symmetry of the collapse. However, the load generated by the jet is a minor precursor to the major loads which occur at the time of complete collapse of the bubble.

This paper discusses thermal analysis in support of probabilistic risk assessment (PRA) to predict the heating of cargoes shipped in vehicles like the Safe Secure Trailer. Fire environments contribute very significantly to the risk associated with ground transport of special nuclear materials. The tradeoff between thermal model complexity and the affordable number of scenarios used to represent the hazard space is discussed as it impacts PRA. The relevant heat transfer mechanisms are discussed along with the applicability of methods from the literature for analysis of these mechanisms. Many of the subject`s real problems remain too complex for affordable and rigorous analysis. Available models are generally restricted to idealizations that are quickly obviated by real effects. Approximate treatment methods, striving to produce conservative, realistic estimates are also discussed.

The suitability of modified thermal-battery technology for use as a potential power source for geothermal borehole applications is under investigation. As a first step, the discharge processes that take place in LiSi/LiBr-KBr-LiF/FeS{sub 2} thermal cells were studied at temperatures of 350 C and 400 C using pelletized cells with immobilized electrolyte. Incorporation of a reference electrode allowed the relative contribution of each electrode to the overall cell polarization to be determined. The results of single-cell tests are presented, along with preliminary data for cells based on a lower-melting CsBr-LiBr-KBr eutectic salt.

An array of chemically-sensitive field-effect transistors (CHEMFETs) that measure both work function and bulk resistance changes in thin films was used to detect volatile organic compounds. Carbon black/organic polymer composite films were deposited onto the CHEMFETs using an automated microdispensing method.

The Waste Isolation Pilot Plant (WIPP) is a planned geologic repository for permanent disposal of transuranic waste generated by US government defense programs. Located near Carlsbad in southeastern New Mexico, the facility`s disposal regions are mined from the bedded salt of the Salado Formation at a depth of approximately 652 m. Four shafts service the operational needs of the facility for air intake, exhaust, waste handling and salt handling. These shafts range in diameter from 3.5 to 6.1 m and extend from the ground surface to the repository. During repository closure, following an operational life of approximately 50 years, these shafts will be sealed in accordance with an acceptable design. Under contract to the US Department of Energy (DOE), the Repository Isolation Systems Department (RISD) of Sandia National Laboratories has developed a design for the WIPP shaft sealing system. This design has been reviewed by the US Environmental Protection Agency (EPA) as part of the 1996 WIPP Compliance Certification Application (CCA). An effective shaft sealing system for the WIPP will limit liquid and gas flows, and permanently prevent the migration of radiological or other hazardous constituents through the sealed shafts from repository to accessible environment. Because of these performance objectives, a significant effort has been directed toward evaluation of the seal design. Whereas RISD (1996) provides a comprehensive discussion, this paper focuses on only one aspect of the evaluation effort, namely a full shaft, fluid flow model.

The `public outcomes` from research universities are educated students and research that extends the frontiers of knowledge. Measures of these `public outcomes` are inadequate to permit either research or education consumers to select research universities based on quantitative performance data. Research universities annually spend over $20 billion on research; 60% of these funds are provided by Federal sources. Federal funding for university research has recently grown at an annual rate near 6% during a time period when other performers of Federal research have experienced real funding cuts. Ten universities receive about 25% of the Federal funds spent on university research. Numerous studies of US research universities are reporting storm clouds. Concerns include balancing research and teaching, the narrow focus of engineering education, college costs, continuing education, and public funding of foreign student education. The absence of research on the `public outcomes` from university research results in opinion, politics, and mythology forming the basis of too many decisions. Therefore, the authors recommend studies of other nations` research universities, studies of various economic models of university research, analysis of the peer review process and how well it identifies the most capable research practitioners and at what cost, and studies of research university ownership of intellectual property that can lead to increased `public outcomes` from publicly-funded research performed by research universities. They advocate two practices that could increase the `public outcomes` from university research. These are the development of science roadmaps that link science research to `public outcomes` and `public outcome` metrics. Changes in the university research culture and expanded use of the Internet could also lead to increased `public outcomes`. They recommend the use of tax incentives to encourage companies to develop research partnerships with research universities.

Vapor diffusion in porous media in the presence of its own liquid may be enhanced due to pore-scale processes, such as condensation and evaporation across isolated liquid islands. Webb and Ho (1997) developed a mechanistic pore-scale model of these processes under steady-state conditions in which condensation and evaporation on the liquid island were equal. The vapor diffusion rate was significantly enhanced by these liquid island processes by up to an order of magnitude compared to a dry porous media. However, vapor transport by diffusion is often complicated by transient effects, such as in drying applications, in which net evaporation of liquid may further augment the vapor flux from diffusion. The influence of transient effects on the enhancement factors for vapor diffusion is evaluated in this paper. In addition, the effect of vapor pressure lowering on the enhancement factor and on porescale vapor fluxes is shown.

The magnetically controlled plasma opening switch (MCPOS) is an advanced plasma opening switch that utilizes magnetic fields to improve operation. Magnetic fields always dominate terawatt, pulsed power plasma opening switches. For that reason, the MCPOS uses controlled applied magnetic fields with magnitude comparable to the self-magnetic field of the storage inductor. One applied field holds the plasma in place while energy accumulates in the storage inductor, then another applied field pushes the plasma away from the cathode to allow energy to flow downstream. Over a ten month period, an MCPOS was designed, built, and tested on DECADE Module 2 at Physics International. The peak drive current was 1.8 MA in 250 ns. The output parameters were up to 1 MA into an electron beam load. The radiation temporal pulse width averaged 60 nanoseconds full-width at half-maximum. The peak load voltage ranged from one to two megavolts. The experiments demonstrated efficient power flow through a long, low-impedance magnetically insulated transmission line between the magnetically controlled plasma opening switch and the load.

Economic and other factors accompanying developments in physics, mathematics and particularly in computer technology are shifting a substantial portion of the experimental resources associated with large scale engineering projects from physical testing to modeling and simulation. In the process, the priorities of selecting meaningful and informative tests and simulations to perform are also changing. This paper describes issues related to experimental design and how the goals and priorities of the experimental design for these problems are changing to accommodate the this shift in experimentation. Issues, priorities and new methods of approach are discussed.

The exposure of ATG graphite sample to DIII-D divertor plasma was provided by the DiMES (Divertor Material Evaluation System) mechanism. The graphite sample arranged to receive the parallel heat flux on a small region of the surface was exposed to 600ms of outer strike point plasma. The sample was constructed to collect the eroded material directed downward into a trapping zone onto s Si disk collector. The average heat flux onto the graphite sample during the exposure was about 200W/cm{sup 2}, and the parallel heat flux was about 10 KW/cm{sup 2}. After the exposure the graphite sample and Si collector disk were analyzed using SEM, NRA, RBS, Auger spectroscopy. IR and Raman spectroscopy. The thermal desorption was studied also. The deposited coating on graphite sample is amorphous carbon layer. Just upstream of the high heat flux zone the redeposition layer has a globular structure. The deposition layer on Si disk is composed also from carbon but has a diamond-like structure. The areal density of C and D in the deposited layer on Si disk varied in poloidal and toroidal directions. The maximum D/C areal density ratio is about 0.23, maximum carbon density is about 3.8 {times} 10{sup 18}cm{sup {minus}2}, maximum D area density is about 3 {times} 10{sup 17}cm{sup 2}. The thermal desorption spectrum had a peak at 1,250K.

Recent observations have indicated that lithium pellet injection wall conditioning plays an important role in achieving the enhanced supershot regime in TFTR. However, little is understood about the behavior of lithium-coated limiter walls, interacting with edge plasmas. In the final campaign of TFTR, a cylindrical carbon fiber composite probe was inserted into the boundary plasma region and exposed to ohmically-heated deuterium discharges with lithium pellet injection. The ion-drift side probe surface exhibits a sign of codeposition of lithium, carbon, oxygen, and deuterium, whereas the electron side essentially indicates high-temperature erosion. It is found that lithium is incorporated in these codeposits in the form of oxide at the concentration of a few percent. In the electron side, lithium has been found to penetrate deeply into the probe material, presumably via rapid diffusion through interplane spaces in the graphite crystalline. Though it is not conclusive, materials mixing in the carbon and lithium system appears to be a key process in successful lithium wall conditioning.

The advanced networking department at Sandia National Laboratories has used the annual Supercomputing conference sponsored by the IEEE and ACM for the past several years as a forum to demonstrate and focus communication and networking developments. At SC `97, Sandia National Laboratories (SNL), Los Alamos National Laboratory (LANL), and Lawrence Livermore National Laboratory (LLNL) combined their SC `97 activities within a single research booth under the Advance Strategic Computing Initiative (ASCI) banner. For the second year in a row, Sandia provided the network design and coordinated the networking activities within the booth. At SC `97, Sandia elected to demonstrate the capability of the Computation Plant, the visualization of scientific data, scalable ATM encryption, and ATM video and telephony capabilities. At SC `97, LLNL demonstrated an application, called RIPTIDE, that also required significant networking resources. The RIPTIDE application had computational visualization and steering capabilities. This paper documents those accomplishments, discusses the details of their implementation, and describes how these demonstrations support Sandia`s overall strategies in ATM networking.

The SAMPLL (Simplified Analytical Model of Penetration with Lateral Loading) computer code was originally developed in 1984 to realistically yet economically predict penetrator/target interactions. Since the code`s inception, its use has spread throughout the conventional and nuclear penetrating weapons community. During the penetrator/target interaction, the resistance of the material being penetrated imparts both lateral and axial loads on the penetrator. These loads cause changes to the penetrator`s motion (kinematics). SAMPLL uses empirically based algorithms, formulated from an extensive experimental data base, to replicate the loads the penetrator experiences during penetration. The lateral loads resulting from angle of attack and trajectory angle of the penetrator are explicitly treated in SAMPLL. The loads are summed and the kinematics calculated at each time step. SAMPLL has been continually improved, and the current version, Version 6.0, can handle cratering and spall effects, multiple target layers, penetrator damage/failure, and complex penetrator shapes. Version 6 uses the latest empirical penetration equations, and also automatically adjusts the penetrability index for certain target layers to account for layer thickness and confinement. This report describes the SAMPLL code, including assumptions and limitations, and includes a user`s guide.