Publications

Structured mesh quality optimization methods are extended to optimization of unstructured triangular, quadrilateral, and mixed finite element meshes. N"ew interpretations of well-known nodally-bssed objective functions are made possible using matrices and matrix norms. The matrix perspective also suggests several new objective functions. Particularly significant is the interpretation of the Oddy metric and the Smoothness objective functions in terms of the condition number of the metric tensor and Jacobian matrix, respectively. Objective functions are grouped according to dimensionality to form weighted combinations. A simple unconstrained local optimum is computed using a modiiied N-ewton iteration. The optimization approach was implemented in the CUBIT mesh generation code and tested on several problems. Results were compared against several standard element-based quaIity measures to demonstrate that good mesh quality can be achieved with nodally-based objective functions.

Capacitor designs for DOE and/or DoD applications are now driven by two major factors; first, the need to reduce component volumes (attain higher energy density) to permit inclusion of additional components and/ or sensors in systems and second, the continuing budget constraints. The reduced volume and cost must be achieved with no sacrifices in functionality, reliability and safety. Since this study was initiated, we have seen a general, continuous increase in resulting short-time breakdown (STB) values, with particular improvements noted on thermal cycled capacitors. Process and results support our prediction that a 50Y0-650A volume reduction can be achieved with no reduction in performance and reliability.

Smoke can cause electronic equipment to fail through increased leakage currents and shorts. Sandia National Laboratories is studying the increased leakage currents caused by smoke with varying characteristics. The objective is to develop models to predict the failure of electronic equipment exposed to smoke. This requires the collection of data on the conductivity of smoke and knowledge of critical electrical systems that control high-consequence operations. We have found that conductivity is a function of the type of fuel, how it is burned, and smoke density. Video recordings of highly biased dc circuits exposed in a test chamber show that during a fire, smoke is attracted to high voltages and can build fragile carbon bridges that conduct leakage currents. The movement of air breaks the bridges, so the conductivity decreases after the fire is extinguished and the test chamber is vented. During the fire, however, electronic equipment may not operate correctly, leading to problems for critical operations dependent on electronic control. The potential for electronic failure is highly dependent on the type of electrical circuit, and Sandia National Laboratories plans to include electrical circuit modeling in the failure models.

When external electric fields are applied to phosphors the cathodoluminescence (CL) at low beam energies is strongly affected. This experiment has been carried out on a variety of common phosphors used in cathode ray tube applications, and the electron beam energy, beam current, and electric field dependence of the CL are thoroughly characterized. It is found that the general features of these effects, particular y the strong polarity and beam energy dependence, are consistent with a model which assumes that the main effect of the electric fields is to alter the populations of electrons `and holes at the phosphor surface. This in turn, modulates the non-radiative energy losses that strongly affect the low-beam-energy CL efficiency. Because the external fields are applied without any direct contact to the phosphor material, the large changes seen in the CL decay rapidly as the beam-created electrons and holes polarize, shielding the externally applied bias. These results have important implications for designing phosphors which might be efficient at low electron energies.

The Solar Two project is a collaborative, cost-shared project between eleven US industry and utility partners and the U.S. Department of Energy to validate the molten-salt power tower technology. The Solar Two plant, located east of Barstow, CA, comprises 1926 heliostats, a receiver, a thermal storage system and a steam generator system that use molten nitrate salt as the heat transfer fluid and storage media. The steam generator powers a 10 MWe, conventional Rankine cycle turbine. This paper describes the test plan and evaluations currently in progress at Solar Two and provides some recent results.

This paper investigates dhlerent strategies that can be used to improve the tracking accuracy of heliostats at Solar Two. The different strategies are analyzed using a geometrical error model to determine their performance over the course of a day. By using the performance of heliostats in representative locations of the field aad on representative days of the year, an estimate of the annual performance of each strategy is presented.

Sucker rod pumps are installed in approximately 90% of all oil wells in the U.S. Although they have been widely used for decades, there are many issues regarding the fluid dynamics of the pump that have not been fully investigated. A project was conducted at Sandia National Laboratories to develop unimproved understanding of the fluid dynamics inside a sucker rod pump. A mathematical flow model was developed to predict pressures in any pump component or an entire pump under single-phase fluid and pumping conditions. Laboratory flow tests were conducted on instrumented individual pump components and on a complete pump to verify and refine the model. The mathematical model was then converted to a Visual Basic program to allow easy input of fluid, geometry and pump parameters and to generate output plots. Examples of issues affecting pump performance investigated with the model include the effects of viscosity, surface roughness, valve design details, plunger and valve pressure differentials, and pumping rate.

Accurate control of the electrode gap in a vacuum arc remelting (VAR) furnace has been a goal of melters for many years. The size of the electrode gap has a direct influence on ingot solidification structure. At the high melting currents (30 to 40 kA) typically used for VAR of segregation insensitive Ti and Zr alloys, process voltage is used as an indicator of electrode gap, whereas drip-short frequency (or period) is usually used at the lower currents (5 to 8 kA) employed during VAR of superalloys. Modem controllers adjust electrode position or drive velocity to maintain a voltage or drip-short frequency (or period) set-point. Because these responses are non-linear functions of electrode gap and melting current, these controllers have a limited range for which the feedback gains are valid. Models are available that relate process voltage and drip-short frequency to electrode gap. These relationships may be used to linearize the controller feedback signal. An estimate of electrode gap may then be obtained by forming a weighted sum of the independent gap estimates obtained from the voltage and drip-short signals. By using multiple independent measures to estimate the gap, a controller that is less susceptible to process disturbances can be developed. Such a controller was designed, built and tested. The tests were carried out at Allvac Corporation during VAR of 12Cr steel at intermediate current levels.

The papers in this special issue have been selected from the systems and balance-of- systems sessions at the 1998 Photovoltaic Performance and Reliability Workshop. The workshop was held November 3-5, 1998 and hosted by the Florida Solar Energy Center, Cocoa Beach, Florida under sponsorship of the US National Center for Photovoltaics (National Renewable Energy Laboratory and Sandia National Laboratories). The topics and issues addressed by these papers were identified in an invited review paper on PV systems by the guest editors. Their work was published earlier this year in Volume 7, Number 1 of Progress in Photovoltaics ('Photovoltaic Systems: An End-of-Millennium Review'). Experts in the PV community were asked to make presentations on these topics at the workshop. The papers that follow are the results of that effort. The papers are organized by topic: (1) codes and standards; (2) reliability; (3) design issues; and (4) commercialization.

Article 690, Solar Photovoltaic Power Systems, has been in the National Electrical Code (NEC) since 1984. An NFPA-appointed Task Group for Article 690 proposed changes to Article 690 for both the 1996 and 1999 codes. The Task Group, supported by more than 50 professionals from throughout the photovoltaic (PV) industry, met seven times during the 1999 code cycle to integrate the needs of the industry with the needs of electrical inspectors and end users to ensure the safety of PV systems. The Task Group proposed 57 changes to Article 690, and all the changes were accepted in the review process. The performance and cost of PV installations were always a consideration as these changes were formed but safety was the number-one priority. All of the proposals were well substantiated and coordinated throughout the PV industry and with representatives of Underwriters Laboratories, Inc (UL). The most significant changes that were made in Article 690 for the 1999 NEC along with some of the rationale are discussed in the remainder of this article.

We have developed instrumentation to enable the combination of surface acoustic wave (SAW) sensor measurements with direct, in-situ molecular spectroscopic measurements to understand the response of the SAW sensors with respect to the interfacial chemistry of surface-confined sensing films interacting with gas-phase analytes. Specifically, the instrumentation and software was developed to perform in-situ Fourier-transform infrared external-reflectance spectroscopy (FTIR-ERS) on operating SAW devices during dosing of their chemically modified surfaces with analytes. By probing the surface with IR spectroscopy during gas exposure, it is possible to understand in unprecedented detail the interaction processes between the sorptive SAW coatings and the gaseous analyte molecules. In this report, we provide details of this measurement system, and also demonstrate the utility of these combined measurements by characterizing the SAW and FTIR-ERS responses of organic thin-film sensor coatings interacting with gas-phase analytes.

Arrays of unheated chemically sensitive resistors (chemiresistors) can serve as extremely small, low-power-consumption sensors with simple read-out electronics. We report here results on carbon-loaded polymer composites, as well as polymeric ionic conductors, as chemiresistor sensors. We use the volubility parameter concept to understand and categorize the chemiresistor responses and, in particular, we compare chemiresistors fabricated from polyisobutylene (PIB) to results from PIB-coated acoustic wave sensors. One goal is to examine the possibility that a small number of diverse chemiresistors can sense all possible solvents-the "Universal Solvent Sensor Array". keywords: chemiresistor, volubility parameter, chemical sensor

The effects of nitrided SiO{sub 2}/Si(100) interfaces upon cycling endurance in silicon-oxide-nitride-oxide-silicon (SONOS) non-volatile memory transistors are investigated. Analysis of MOSFET sub-threshold characteristics indicate cycling degradation to be a manifestation of interface state (D{sub it}) generation at the tunnel oxide/silicon interface. After 10{sup 6} write/erase cycles, SONOS film stacks prepared with nitrided tunnel oxides exhibit enhanced cycling endurance with {Delta}D{sub it}=3x10{sup 12} V{sup -1}cm{sup -2}, compared to {Delta}D{sub it}=2x10{sup 13} V{sup -l}cm{sup -2} for non-nitrided tunnel oxides. Additionally, if the capping oxide is formed by steam oxidation, rather than by deposition, SONOS stacks prepared with non-nitrided tunnel oxides exhibit endurance characteristics similar to stacks with nitrided tunnel oxides. From this observation it is concluded that latent nitridation of the tunnel oxidehilicon interface occurs during steam oxide cap formation.

Growth kinetics, mechanisms, and material quality in GaN epitaxial lateral over-growth (ELO) were examined using a single mask of systematically varied patterns. A 2-D gas phase reaction/diffusion model describes how transport of the Ga precursor to the growth surface enhances the lateral rate in the early stages of growth. In agreement with SEM studies of truncated growth runs, the model also predicts the dramatic decrease in the lateral rate that occurs as GaN over-growth reduces the exposed area of the mask. At the point of convergence, a step-flow coalescence mechanism is observed to fill in the area between lateral growth-fronts. This alternative growth mode in which a secondary growth of GaN is nucleated along a single convergence line, may be responsible for producing smooth films observed to have uniform cathodoluminescence (CL) when using 1{micro}m nucleation zones. Although emission is comprised of both UV ({approximately}365nm) and yellow ({approximately}550nm) components, the spectra suggest these films have reduced concentrations of threading dislocations normally associated with non-radiative recombination centers and defects known to accompany growth-front convergence lines.

The organization and assembly of molecules in cellular membranes is orchestrated through the recognition and binding of specific chemical signals. A simplified version of the cellular membrane system has been developed using a synthetically prepared membrane receptor incorporated into a biologically derived lipid bilayer. Through an interplay of electrostatic and van der Wards interactions, aggregation or dispersion of molecular components could be executed on command using a specific chemical signal. A pyrene fluorophore was used as an optical probe to monitor the aggregational state of the membrane receptors in the bilayer matrix. The pyrene excimer emission to monomer emission (E/M) intensity ratio gave a relative assessment of the local concentration of receptors in the membrane. Bilayers were prepared with receptors selective for the divalent metal ions of copper, mercury, and lead. Addition of the metal ions produced a rapid dispersion of aggregated receptor components at nano- to micro-molar concentrations. The process was reversible by sequestering the metal ions with EDTA. Receptors for proteins and polyhistidine were also prepared and incorporated into phosphatidylcholine lipid bilayers. In this case, the guest molecules bound to the membrane through multiple points of interaction causing aggregation of initially dispersed receptor molecules. The rapid, selective, and sensitive fluorescence optical response of these lipid assemblies make them attractive in sensor applications for aqueous phase metal ions and polypeptides.

The National Electrical Code@ (NEC@) focuses primarily on electrical system installation requirements in the U.S. The NEC addresses both fire and personnel safety. This paper will describe recent efforts of the PV industry in the U.S. and the resulting requirements in the 1999 National Electrical Code-- Article 690 --Solar Photovoltaic Systems. The Article 690 requirements spell out the PV-unique requirements for safe installations of PV systems in the U.S.A. This paper provides an overview of the most significant changes that appear in Article 690 of the 1999 edition of the NEC. The related and coordinated efforts of the other standards- making groups will also be briefly reviewed.

Using surface micromachined samples, we demonstrate the accurate measurement of cantilever beam adhesion by using test structures which are adhered over long attachment lengths. We show that this configuration has a deep energy well, such that a fracture equilibrium is easily reached. When compared to the commonly used method of determining the shortest attached beam, the present method is much less sensitive to variations in surface topography or to details of capillary drying.

Abstract not provided.

Over the past decade, Sandia National Laboratories has been involved in the development of receivers to transfer energy from the focus of a parabolic dish concentrator to the heater tubes of a Stirling engine. Through the isothermal evaporation and condensation of sodium. a heat-pipe receiver can efficiently transfer energy to an engine's working fluid and compensate for irregularities in the flux distribution that is delivered by the concentrator. The operation of the heat pipe is completely passive because the liquid sodium is distributed over the solar-heated surface by capillary pumping provided by a wick structure. Tests have shown that using a heat pipe can boost the system performance by twenty percent when compared to directly illuminating the engine heater tubes. Designing heat pipe solar receivers has presented several challenges. The relatively large area ({approximately}0.2 m{sup 2}) of the receiver surface makes it difficult to design a wick that can continuously provide liquid sodium to all regions of the heated surface. Selecting a wick structure with smaller pores will improve capillary pumping capabilities of the wick, but the small pores will restrict the flow of liquid and generate high pressure drops. Selecting a wick that is comprised of very tine filaments can increase the permeability of the wick and thereby reduce flow losses, however, the fine wick structure is more susceptible to corrosion and mechanical damage. This paper provides a comprehensive review of the issues encountered in the design of heat pipe solar receivers and solutions to problems that have arisen. Topics include: flow characterization in the receiver, the design of wick systems. the minimization of corrosion and dissolution of metals in sodium systems. and the prevention of mechanical failure in high porosity wick structures.

High integrity/high consequence systems must be safe and reliable; hence it is only logical that both software safety and software reliability cases should be developed. Risk assessments in safety cases evaluate the severity of the consequences of a hazard and the likelihood of it occurring. The likelihood is directly related to system and software reliability predictions. Software reliability cases, as promoted by SAE JA 1002 and 1003, provide a practical approach to bridge the gap between hardware reliability, software reliability, and system safety and reliability by using a common methodology and information structure. They also facilitate early insight into whether or not a project is on track for meeting stated safety and reliability goals, while facilitating an informed assessment by regulatory and/or contractual authorities.

The Boeing Company's Dish Engine Critical Component (DECC) project started in April of 1998. It is a continuation of a solar energy program started by McDonnell Douglas (now Boeing) and United Stirling of Sweden in the mid 1980s. The overall objectives, schedule, and status of this project are presented in this paper. The hardware test configuration, hardware background, operation, and test plans are also discussed. A summary is given of the test data, which includes the daily power performance, generated energy, working-gas usage, mirror reflectivity, solar insolation, on-sun track time, generating time, and system availability. The system performance based upon the present test data is compared to test data from the 1984/88 McDonnell Douglas/United Stirling AB/Southem California Edison test program. The test data shows that the present power, energy, and mirror performance is comparable to when the hardware was first manufactured 14 years ago.

Motion detectors consisting of Pb(Zr{sub x}Ti{sub (1{minus}x)})O{sub 3} (PZT) thin films, between platinum electrodes, on micromachined silicon compound clamped-clamped or cantilever beam structures were fabricated using either hot KOH or High Aspect Ratio Silicon Etching (HARSE) to micromachine the silicon. The beams were designed such that a thicker region served as a test mass that produced stress at the top of the membrane springs that supported it when the object to which the detector was mounted moved. The PZT film devices were placed on these membranes to generate a charge or a voltage in response to the stress through the piezoelectric effect. Issues of integration of the PZT device fabrication process with the two etching processes are discussed. The effects of PZT composition and device geometry on the response of the detectors to motion is reported and discussed.

We have developed a variety of processes for fabricating components for micro devices based on deep x-ray lithography (DXRL). Although the techniques are applicable to many materials, we have demonstrated them using hard (Nd{sub 2}Fe{sub 14}B) and soft (Ni-Zn ferrite) magnetic materials because of the importance of these materials in magnetic micro-actuators and other devices and because of the difficulty fabricating them by other means. The simplest technique involves pressing a mixture of magnetic powder and a binder into a DXRL-formed mold. In the second technique, powder is pressed into the mold and then sintered to densify. The other two processes involve pressing at high temperature either powder or a dense bulk material into a ceramic mold that was previously made using a DXRL mold. These techniques allow arbitrary 2-dimensional shapes to be made 10 to 1000 micrometers thick with in-plane dimensions as small as 50 micrometers and dimensional tolerances in the micron range. Bonded isotropic Nd{sub 2}Fe{sub 14}B micromagnets made by these processes had an energy product of 7 MGOe.

This Institutional Plan is the most comprehensive yearly "snapshot" available of Sandia National Laboratories' major programs, facilities, human resources, and budget. The document also includes overviews of our missions, organization, capabilities, planning functions, milestones, and accomplishments. The document's purpose is to provide the above information to the US Department of Energy, key congressional committees, Sandia management, and other present and potential customers. Chapter 2 presents information about Sandia's mission and summarizes our recent revision of Sandia's Strategic Plan. Chapter 3 presents an overview of Sandia's strategic objectives, chapter 4 lists laboratory goals and milestones for FY 1999, and chapter 5 presents our accomplishments during FY 1998. Chapters 3 through 5 are organized around our eight strategic objectives. The four primary objectives cover nuclear weapons responsibilities, nonproliferation and materials control, energy and critical infrastructures, and emerging national security threats. The major programmatic initiatives are presented in chapter 7. However, the programmatic descriptions in chapter 6 and the Associated funding tables in chapter 9 continue to be presented by DOE Budget and Reporting Code, as in previous Sandia institutional plans. As an aid to the reader, the four primary strategic objectives in chapter 3 are cross-referenced to the program information in chapter 6.

Mining has always had an important influence on cultures and traditions of communities around the globe and throughout history. Today, because mining legislation places heavy emphasis on environmental protection, there is great interest in having a comprehensive understanding of ancient mining and mining sites. Multi-disciplinary approaches (i.e., Pb isotopes as tracers) are being used to explore the distribution of metals in natural environments. Another successful approach is to model solution migration numerically. A proven method to simulate solution migration in natural rock salt has been applied to project through time for 10,000 years the system performance and solution concentrations surrounding a proposed nuclear waste repository. This capability is readily adaptable to simulate solution migration around mining.