Publications

The solution of Grand Challenge Problems will require computations that are too large to fit in the memories of even the largest machines. Inevitably, new designs of I/O systems will be necessary to support them. This report describes the work in investigating I/O subsystems for massively parallel computers. Specifically, the authors investigated out-of-core algorithms for common scientific calculations present several theoretical results. They also describe several approaches to parallel I/O, including partitioned secondary storage and choreographed I/O, and the implications of each to massively parallel computing.

This contribution discusses why, and how, mobile networks and mobile switches might be discussed during Phase 1 of the WATM standards process. Next, it reviews mobile routers within Mobile IP. That IP mobility architecture may not apply to the proposed mobile ATM switches. Finally, it discusses problems with PNNI peer group formation and operation when mobile ATM switches are present.

Crystalline silicotitanates (CSTs) are a new class of ion exchangers that were jointly invented by researchers at Sandia National Laboratories and Texas A&M University. One particular CST, known as TAM-5, is remarkable for its ability to separate parts-per-million concentrations of cesium from highly alkaline solutions (pH> 14) containing high sodium concentrations (>5M). It is also highly effective for removing cesium from neutral and acidic solutions, and for removing strontium from basic and neutral solutions. Cesium isotopes are fission products that account for a large portion of the radioactivity in waste streams generated during weapons material production. Tests performed at numerous locations with early lab-scale TAM-5 samples established the material as a leading candidate for treating radioactive waste volumes such as those found at the Hanford site in Washington. Thus Sandia developed a Cooperative Research and Development Agreement (CRADA) partnership with UOP, a world leader in developing, commercializing, and supplying adsorbents and associated process technology to commercialize and further develop the material. CSTs are now commercially available from UOP in a powder (UOP IONSIV{reg_sign} IE-910 ion exchanger) and granular form suitable for column ion exchange operations (UOP IONSIV{reg_sign} IE-911 ion exchanger). These materials exhibit a high capacity for cesium in a wide variety of solutions of interest to the Department of Energy, and they are chemically, thermally, and radiation stable. They have performed well in tests at numerous sites with actual radioactive waste solutions, and are being demonstrated in the 100,000 liter Cesium Removal Demonstration taking place at Oak Ridge National Laboratory with Melton Valley Storage Tank waste. It has been estimated that applying CSTs to the Hanford cleanup alone will result in a savings of more than $300 million over baseline technologies.

This paper presents the design, results, and analysis of a high-brightness electron beam technology demonstration experiment completed at Sandia National Laboratories, performed in collaboration with Los Alamos National Laboratory. The anticipated electron beam parameters were: 12 MeV, 35-40 kA, 0.5-mm rms radius, and 40-ns full width half maximum (FWHM) pulse duration. This beam, on an optimum thickness tantalum converter, should produce a very intense x-ray source of {approximately} 1.5-mm spot size and 1 kR dose @ 1 m. The accelerator utilized was SABRE, a pulsed inductive voltage adder, and the electron source was a magnetically immersed foilless electron diode. For these experiments, SABRE was modified to high-impedance negative-polarity operation. A new 100-ohm magnetically insulated transmission line cathode electrode was designed and constructed; the cavities were rotated 180{degrees} poloidally to invert the central electrode polarity to negative; and only one of the two pulse forming lines per cavity was energized. A twenty- to thirty-Tesla solenoidal magnet insulated the diode and contained the beam at its extremely small size. These experiments were designed to demonstrate high electron currents in submillimeter radius beams resulting in a high-brightness high-intensity flash x-ray source for high-resolution thick-object hydrodynamic radiography. The SABRE facility high-impedance performance was less than what was hoped. The modifications resulted in a lower amplitude (9 MV), narrower-than-anticipated triangular voltage pulse, which limited the dose to {approximately} 20% of the expected value. In addition, halo and ion-hose instabilities increased the electron beam spot size to > 1.5 mm. Subsequent, more detailed calculations explain these reduced output parameters. An accelerator designed (versus retrofit) for this purpose would provide the desired voltage and pulse shape.

Current robotic systems are unable to recognize most unexpected changes in the work environment, such as tool breakage, workpiece motion, or sensor failure. Unless halted by a human operator, they are likely to continue actions that are at best inappropriate, and at worst may cause damage to the workpiece or robot. This project investigated use of Artificial Neural Networks (ANNs) to learn the expected characteristics of sensor data during normal operations, recognize when data no longer is consistent with normal operation, suspend operations and alert a human operator. Data on force and torque applied at the robot tool tip were collected from two workcells: a robotic deburring system and a robot material-handling system. Data were collected for normal operations and for operations in which a fault condition was introduced. Data simulating sensor failure and excessive sensor noise were generated. Artificial Neural Networks (ANN) were trained to classify operating conditions; several ANN architectures were tested. The selected ANNs were able to correctly classify all valid operating conditions and the majority of fault conditions over the entire range of operating conditions, having {open_quotes}learned{close_quotes} the expected force/torque data. Most faults introduced appreciable error in the data; these were correctly classified. However, a small minority of faults did not give rise to a detectable difference in force and torque data. It is believed that these faults could be detected using other sensors. The computational workload varies with the implementation, but is moderate: up to 2.3 megaflops. This makes implementation of a real-time workcell monitor feasible.

Aerospace components are often subjected to pyroshock events during flight and deployment, and must be qualified to this frequently severe environment. Laboratory simulation of pyroshock using a mechanically excited resonant fixture, has gained favor at Sandia for testing small (<8 inch cube) weapon components. With this method, each different shock environment required a different resonant fixture that was designed such that it`s response matched the environment. In Phase 1 (SAND92-2135) of this research, a new test method was developed which eliminated the need to have a different resonant fixture for each test requirement. This was accomplished by means of a tunable resonant fixture that has a response which is adjustable over a wide frequency range. The adjustment of the fixture`s response is done in a simple and deterministic way. This report covers Phase 2 of this research, in which several ideas were explored to extend the Phase 1 results to a larger scale. The test apparatus developed in Phase 1 was capable of testing components with up to a 10 inches x 10 inches base. The goal of the Phase 2 research was to produce an apparatus capable of testing components with up to a 20 inches x 20 inches mounting base. This size capability would allow the testing of most satellite and missile components which frequently consist of large electronic boxes. Several methods to attain this goal were examined, including scaling up the Phase 1 apparatus. Only one of these proved capable of meeting the Phase 2 goals. This report covers all details from concept through fabrication and testing of this Phase 2 apparatus.

Structural system identification is concerned with the development of systematic procedures and tools for developing predictive analytical models based on a physical structure`s dynamic response characteristics. It is a multidisciplinary process that involves the ability (1) to define high fidelity physics-based analysis models, (2) to acquire accurate test-derived information for physical specimens using diagnostic experiments, (3) to validate the numerical simulation model by reconciling differences that inevitably exist between the analysis model and the experimental data, and (4) to quantify uncertainties in the final system models and subsequent numerical simulations. The goal of this project was to develop structural system identification techniques and software suitable for both research and production applications in code and model validation.

Viscoelastic materials are often characterized in terms of stress relaxation moduli which decay in time. Finite element programs which model viscoelastic materials frequently require that these relaxation functions be defined as an exponential series (i.e., Prony Series) to exploit the numerical advantages of developing recursive equations for evaluating hereditary integrals. Obtaining these data fits can be extremely difficult when the data is spread over many decades in the logarithm of time. RELFIT is a nonlinear optimization program that iteratively determines the Prony series coefficients and relaxation times so as to minimize the least squares error in the data fit. An overview of the code, a description of the required inputs (i.e., users`s instructions), and a demonstration problem are presented.

This report summarizes research completed under a Laboratory Directed Research and Development program funded for part of FY94, FY95 and FY96. The main goals were (1) to develop novel, compound-semiconductor based optical sources to enable field-based detection of environmentally important chemical species using miniaturized, low-power, rugged, moderate cost spectroscopic equipment, and (2) to demonstrate the utility of near-infrared spectroscopy to quantitatively measure contaminants. Potential applications would include monitoring process and effluent streams for volatile organic compound detection and sensing head-space gasses in storage vessels for waste management. Sensing is based on absorption in the 1.3-1.9 {mu}m band from overtones of the C-H, N-H and O-H stretch resonances. We describe work in developing novel broadband light-emitting diodes emitting over the entire 1.4-1.9 {mu}m wavelength range, first using InGaAs quantum wells, and second using a novel technique for growing digital-alloy materials in the InAlGaAs material system. Next we demonstrate the utility of near-infrared spectroscopy for quantitatively determining contamination of soil by motor oil. Finally we discuss the separability of different classes of organic compounds using near-infrared spectroscopic techniques.

This report summarizes the purchasing and transportation activities of the Procurement Organization for Fiscal Year 1996, Activities for both the New Mexico and California locations are included.

The multifrequency, multisource integral wave migration method commonly used in the analysis of seismic data is extended to electromagnetic (EM) data within the audio frequency range. The method is applied to the secondary magnetic fields produced by a borehole, vertical electric source (VES). The integral wave-migration method is a numerical reconstruction procedure utilizing Green`s theorem where the fields are migrated (extrapolated) from the measuring aperture into the interior of the earth. To form the image, the approach used here is to Fourier transform the constructed image from frequency domain to time domain and set time equal to zero. The image is formed when the in-phase part (real part) is a maximum or the out-of-phase (imaginary part) is a minimum; ie., the EM wave is phase coherent at its origination. In the application here, the secondary magnetic fields are treated as scattered fields. To determine the conductivity, the measured data migrated to a pixel location are equated to calculated data migrated to the same pixel. The conductivity is determined from solving a Fredholm integral equation of the first kind by solving a system of linear algebraic equations. The multifrequency, multisource integral wave-migration method is applied to calculated model data and to actual field data acquired to map a diesel fuel oil spill. For the application discussed here, a two dimensional resistivity slice is calculated from the solution to the Fredholm integral equation. The resistivity image of the fuel oil agrees with the known location.

This project utilized Computer-Aided Molecular Design (CAMD) to develop a new class of metalloporphyrin materials for use as catalysts for two fuel cell reactions. The first reaction is the reduction of oxygen at the fuel cell cathode, and this reaction was the main focus of the research. The second reaction we attempted to catalyze was the oxidation of methanol at the anode. Two classes of novel metalloporphyrins were developed. The first class comprised the dodecaphenylporphyrins whose steric bulk forces them into a non-planar geometry having a pocket where oxygen or methanol is more tightly bound to the porphyrin than it is in the case of planar porphyrins. Significant improvements in the catalytic reduction of oxygen by the dodecaphenyl porphyrins were measured in electrochemical cells. The dodecaphenylporphyrins were further modified by fluorinating the peripheral phenyl groups to varying degrees. The fluorination strongly affected their redox potential, but no effect on their catalytic activity towards oxygen was observed. The second class of porphyrin catalysts was a series of hydrogen-bonding porphyrins whose interaction with oxygen is enhanced. Enhancements in the interaction of oxygen with the porphyrins having hydrogen bonding groups were observed spectroscopically. Computer modeling was performed using Molecular Simulations new CERIUS2 Version 1.6 and a research version of POLYGRAF from Bill Goddard`s research group at the California Institute of Technology. We reoptimized the force field because of an error that was in POLYGRAF and corrected a problem in treatment of the metal in early versions of the program. This improved force field was reported in a J. Am. Chem. Soc. manuscript. Experimental measurements made on the newly developed catalysts included the electrochemical testing in a fuel cell configuration and spectroscopic measurements (UV-Vis, Raman and XPS) to characterize the catalysts.

A magnetostrictive borehole seismic source was developed for use in high resolution crosswell surveys in environmental applications. The source is a clamped, vertical-shear, swept frequency, reaction-mass shaker design consisting of a spring pre-loaded magnetostrictive rod with permanent magnet bias, drive coils to induce an alternating magnetic field, and an integral tungsten reaction mass. The actuator was tested extensively in the laboratory. It was then incorporated into an easily deployable clamped downhole tool capable of operating on a standard 7 conductor wireline in borehole environments to 10,000{degrees} deep and 100{degrees}C. It can be used in either PVC or steel cased wells and the wells can be dry or fluid filled. It has a usable frequency spectrum of {approx} 150 to 2000 Hz. The finished tool was successfully demonstrated in a crosswell test at a shallow environmental site at Hanford, Washington. The source transmitted signals with a S/N ratio of 10-15 dB from 150-720 Hz between wells spaced 239 feet apart in unconsolidated gravel. The source was also tested successfully in rock at an oil field test site, transmitting signals with a S/N ratio of 5-15 dB over the full sweep spectrum from 150-2000 Hz between wells spaced 282 feet apart. And it was used successfully on an 11,000{degrees} wireline at a depth of 4550{degrees}. Recommendations for follow-on work include improvements to the clamp, incorporation of a higher sample rate force feedback controller, and increases in the force output of the tool.

The Telemetry Technology Development Department have, in support of the Advanced Geophysical Technology Department and the Oil Recovery Technology Partnership, developed a fiber optic communication capability for use in borehole applications. This environment requires the use of packaging and component technologies to operate at high temperature (up to 175{degrees}C) and survive rugged handling. Fiber optic wireline technology has been developed by The Rochester Corporation under contract to Sandia National Labs and produced a very rugged, versatile wireline cable. This development has utilized commercial fiber optic component technologies and demonstrated their utility in extreme operating environments.

Carbon nanotubes and their composites were examined using computational and experimental techniques in order to modify the mechanical and electrical properties of resins. Single walled nanotubes were the focus of the first year effort; however, sufficient quantities of high purity single walled nanotubes could not be obtained for mechanical property investigations. The unusually high electrical conductivity of composites loaded with <1% of multiwalled nanotubes is useful, and is the focus of continuing, externally funded, research.

A technology roadmap is the result of a strategic technology planning process that cooperatively identifies (1) a particular industry`s common product and process performance targets, (2) the technology alternatives and milestones for meeting these targets, and (3) a common technology path for research and development activities. The author describes a successful major roadmapping experience - the Semiconductor Industry Association`s Technology Roadmapping Process, which culminated in a workshop held in 1992. The report explains the committee structure and processes that were used both before and after the workshop and presents principles and practices that can aid future technology roadmappers. Appendix 1 summarizes the process from a committee-structure viewpoint. Appendix 2 summarizes the process from a functional view-point. Appendix 3 answers some frequently asked questions about technology roadmapping.

This paper presents a new method for wideband force balancing a proof-mass in multiple axes simultaneously. Capacitive position sense and force feedback are accomplished using the same air-gap capacitors through time multiplexing. Proof of concept is experimentally demonstrated with a single-mass monolithic surface micromachined 3-axis accelerometer.

The ability to successfully use and interact with a computerized world model is dependent on the ability to create an accurate world model. The goal of this project was to develop a prototype system to remotely deploy sensors into a workspace, collect surface information, and rapidly build an accurate world model of that workspace. A key consideration was that the workspace areas are typically hazardous environments, where it is difficult or impossible for humans to enter. Therefore, the system needed to be fully remote, with no external connections. To accomplish this goal, an electric, mobile platform with battery power sufficient for both the platform and sensor electronics was procured and 3D range sensors were deployed on the platform to capture surface data within the workspace. A radio Ethernet connection was used to provide communications to the vehicle and all on-board electronics. Video from on-board cameras was also transmitted to the base station and used to teleoperate the vehicle. Range data generated by the on-board 3D sensors was transformed into surface maps, or models. Registering the sensor location to a consistent reference frame as the platform moved through the workspace allowed construction of a detailed 3D world model of the extended workspace.

Technology planning is important for many reasons. Globally, companies are facing many competitive problems. Technology roadmapping, a form of technology planning can help deal with this increasingly competitive environment. While it has been used by some companies and industries, the focus has always been on the technology roadmap as a product, not on the process. This report focuses on formalizing the process so that it can be more broadly and easily used. As a DOE national security laboratory with R&D as a major product, Sandia must do effective technology planning to identify and develop the technologies required to meet its national security mission. Once identified, technology enhancements or new technologies may be developed internally or collaboratively with external partners. For either approach, technology roadmapping, as described in this report, is an effective tool for technology planning and coordination, which fits within a broader set of planning activities. This report, the second in a series on technology roadmapping, develops and documents this technology roadmapping process, which can be used by Sandia, other national labs, universities, and industry. The main benefit of technology roadmapping is that it provides information to make better technology investment decisions by identifying critical technologies and technology gaps and identifying ways to leverage R&D investments. It can also be used as a marketing tool. Technology roadmapping is critical when the technology investment decision is not straight forward. This occurs when it is not clear which alternative to pursue, how quickly the technology is needed, or when there is a need to coordinate the development of multiple technologies. The technology roadmapping process consists of three phases - preliminary activity, development of the technology roadmap, and follow-up activity.

Since micromachining technology has raised the prospect of fabricating high performance sensors without the associated high cost and large size, many researchers have investigated micromachined rate gyroscopes. The vast majority of research has focused on single input axis rate gyroscopes, but this paper presents work on a dual input axis micromachined rate gyroscope. The key to successful simultaneous dual axis operation is the quad symmetry of the circular oscillating rotor design. Untuned gyroscopes with mismatched modes yielded random walk as low as 10{degrees}/{radical}hour with cross sensitivity ranging from 6% to 16%. Mode frequency matching via electrostatic tuning allowed performance better than 2{degrees}/{radical}hour, but at the expense of excessive cross sensitivity.

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 Supercomputing 96, for the first time, Sandia National Laboratories, Los Alamos National Laboratory, and Lawrence Livermore National Laboratory combined their Supercomputing 96 activities within a single research booth under the ASO banner. Sandia provided the network design and coordinated the networking activities within the booth. At Supercomputing 96, Sandia elected: to demonstrate wide area network connected Massively Parallel Processors, to demonstrate the functionality and capability of Sandia`s new edge architecture, to demonstrate inter-continental collaboration tools, and to demonstrate ATM video 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.

Ross developed an analytical relationship to calculate the diversion length of a tilted fine-over-coarse capillary barrier. Oldenburg and Pruess compared simulation results using upstream and harmonic weighting to the diversion length predicted by Ross formula with mixed results; the qualitative agreement is reasonable but the quantitative comparison is poor, especially for upstream weighting. The proximity of the water table to the fine-coarse interface at breakthrough is a possible reason for the poor agreement. In the present study, the Oldenburg and Pruess problem is extended to address the water table issue. When the water table is sufficiently far away from the interface at breakthrough, good qualitative and quantitative agreement is obtained using upstream weighting.

This paper reviews how elements from the theory of fractal functions are employed to construct scaling vectors and multiwavelets. Emphasis is placed on the one-dimensional case, however extensions to IR{sup m} are indicated.

The project addressed the need of the oil and gas industry for real-time information about the drilling process and the formations being drilled. An ideal system would allow measuring while drilling (MWD) and would transmit data to the surface immediately at a rate high enough to support video or televiewer systems. A proposed solution was to use an optical fiber as a link between the surface and the instrumentation package. We explored the use of a disposable MWD telemetry cable, drawing on the technology developed for missile guidance which deploys miles of fiber from a small spool at missile speeds approaching half the speed of sound. Emphasis In was on the questions of survivability of the unarmored fiber in the drill string environment and deployability. Laboratory and field testing showed the concept worked under realistic conditions; a field demonstration transmitted data at 10 kilobits per second from a depth of 3500 feet.

We are testing an anti-weathering preservation strategy that is specific to limestone surfaces. The strategy involves the application of a mineral-specific, bifunctional, passivating/coupling agent that binds to both the limestone surface and to the consolidating inorganic polymer matrix. The sol-gel based reactions form composite materials with desirable conservation and anti-weathering properties. We present the results of our efforts, the highlights of which are: (1) scanning probe microscopy of moisture-free calcite crystals treated with the trisilanol form of silylalkylaminocarboxylate (SAAC), reveals porous agglomerates that offer no significant resistance to the mild leaching action of deionized water. When the crystals are further consolidated with a silica-based consolidant (A2**), no dissolution is seen although the positive role of the passivant molecule is not yet delineated. (2) Modulus of rupture tests on limestone cores treated with an aminoalkylsilane (AEAPS) and A2** showed a 25-35% increase in strength compared to the untreated samples. (3) Environmental scanning electron microscopy of treated limestone subjected to a concentrated acid attack showed degradation of the surface except in areas where thick layers of the consolidant were deposited.