Publications

An element based finite control volume procedure is applied to the solution of ablation problems for 2-D axisymmetric geometries. A mesh consisting of four node quadrilateral elements was used. The nodes are allowed to move in response to the surface recession rate. The computational domain is divided into a region with a structured mesh with moving nodes and a region with an unstructured mesh with stationary nodes. The mesh is costrained to move along spines associated with the original mesh. Example problems are presented for the ablation of a realistic nose tip geometry exposed to aerodynamic heating from a uniform free stream environment.

The paper gives some of the highlights of a panel discussion on surface diffusion held Monday, November 30, 1992 at the Fall MRS Meeting in Boston, Massachusetts. Four invited speakers discussed computer modeling techniques and scanning tunneling microscopy experiments that have been used to provide new understanding of the atomistic processes that occur at surfaces. We present a summary of each of the invited talks, indicate other presentations on surface diffusion in this proceedings, and provide a transcript of the two discussion sessions.

During the past few years, methods have been developed for quantifying and analyzing common cause failures (CCFs). These methods have outpaced current data collection activities. This document discusses the collection and documentation of failure events at nuclear power plants with respect to these new CCFs methods. The report concentrates on the information necessary to improve the parameter estimates for both independent and dependent events in probabilistic risk assessments (PRAS) and alludes to the fact that the same information can be used to enhance other nuclear power plant activities. Several existing data bases are reviewed as to their adequacy for these new CCF methods, and areas where information is lacking, either because certain information is simply not required to be reported or because required information was simply not reported, are identified. Finally, data needs identified from recent PRAs are discussed.

Lighting protection systems (LPSs) for explosives handling and storage facilities have long been designed similarly to those will for more conventional facilities, but their overall effectiveness in controlling interior electromagnetic (EM) environments has still not been rigorously assessed. Frequent lightning-caused failures of a security system installed in earth-covered explosives storage structures prompted the U.& Army and Sandia National Laboratories to conduct a program to determine quantitatively the EM environments inside an explosives storage structure that is struck by lightning. These environments were measured directly during rocket-triggered lightning (RTL) tests in the summer of 1991 and were computed using linear finite-difference, time-domain (FDTD) EM solvers. The experimental and computational results were first compared in order to validate the code and were also used to construct bounds for interior environments corresponding to seven incident lightning flashes. The code insults were also used to develop simple circuit models for the EM field behavior-a process that insulted in a very simple and somewhat surprising physical interpretation of the structure`s response that has significant practical and economic implications for design, construction, and maintenance of such facilities.

A panel discussion on interface roughness was held at the Fall 1992 Materials Research Society meeting. We present a of results presented by the invited speakers on the application and interpretation of X-ray reflectivity, atomic force microscopy (AFM), scanning tunneling microscopy (STM), photoluminescence and transmission electron microscopy.

A safety system has been designed and constructed to mitigate the asphyxiation and low temperature hazards presented by the distribution and usage of cryogenic liquids in work spaces at Sandia National Laboratories. After identifying common accident scenarios, the CRYOFACS (Cryogenic Fail-Safe Control System) unit was designed, employing microprocessor technology and software that can be easily modified to accommodate varying laboratory requirements. Sensors have been incorporated in the unit for the early detection of accidental releases or overflows of cryogenic liquids. The CRYOFACS design includes control (and shutdown) of the cryogen source upon error detection, and interfaces with existing oxygen monitors, in common use at Sandia Labs, to provide comprehensive protection for both personnel and property.

The Greater Confinement Disposal (GCD) facility was established by the Nevada office of the Department of Energy (DOE) in Area 5 at the Nevada Test Site for containment of waste inappropriate for shallow land burial. Some transuranic (TRU) waste has been disposed of at the GCD facility, and compliance of this disposal system with Environmental Protection Agency (EPA) regulations 40 CFR 191 must be evaluated by performance assessment calculations. We have adopted an iterative approach where performance assessment results guide site data collection which in turn influences the parameters and models used in performance assessment. The first iteration was based upon readily available data. The first iteration indicated that the GCD facility would likely comply with 40 CFR 191 and that the downward recharge rate had a major influence on the results. As a result, a site characterization project was initiated to study recharge in Area 5 by use of three environmental tracers. This study resulted in the conclusion that recharge was extremely small, if not negligible. Thus, downward advection to the water table is no longer considered a viable release pathway, leaving upward liquid diffusion as the sole release pathway. This second performance assessment iteration refined the upward pathway models and parameters. The results of the performance assessment using these models still indicate that the GCD site is likely to comply with all sections of 40 CFR 191.

We have developed a method for generating chromate-free corrosion resistant coatings on aluminum alloys using a process procedurally similar to standard chromate conversion. These coatings provide good corrosion resistance on 6061-T6 and 1100 A1 under salt spray testing conditions. The resistance of the new coating is comparable to that of chromate conversion coatings in four point probe tests, but higher when a mercury probe technique is used. Initial tests of paint adhesion, and under paint corrosion resistance are promising. Primary advantage of this new process is that no hazardous chemicals are used or produced during the coating operation.

A method is described to characterize shocks (transient time histories) in terms of the Fourier energy spectrum and the temporal moments of the shock passed through a contiguous set of bandpass filters. This method is compared for two transient time histories with the more conventional methods of shock response spectra (SRS) and a nonstationary random characteristic.

Qualitatively different trends in postirradiation electrical response are observed in MOS devices after very long (up to 2.75-year) switched-bias bakes. A revised defect nomenclature is introduced, and implications for MOS defect models are discussed.

La{sub 2-x}Sr{sub x}CuO{sub 4+{delta}} with x = 0.01, 0.025, 0.050, 0.10 and 0.16 and excess oxygen {delta} incorporated by high-pressure O{sub 2} anneals. These compounds were examined using time-of-flight neutron diffraction data. Various models were fit by Rietveld least-squares refinement, with the maximum amount of {delta} being only of the order of 10 standard deviations. {delta} is largest for x near 0, is zero for x = 0.10 and is intermediate for x = 0.16. Only the sample with x = 0.01 is found to phase separate distinctly into a nearly stoichiometric phase with {delta} {approx} 0 and an oxygen-rich superconducting phase as the temperature is lowered. Coincidence of phase separation and Neel temperature strongly suggests that the phase separation is driven by free energy provided by long-range antiferromagnetic ordering in the nearly stoichiometric, weakly Sr-doped La{sub 2-x}Sr{sub x}CuO{sub 4}. The excess oxygen stoichiometry shows that at low values of x, hole doping is provided primarily by the excess oxygen, and is enhanced substantially by phase separation. At larger values of x, excess oxygen is no longer incorporated, and hole doping is provided by the substitution of Sr{sup +2} for La{sup +3}.

The US Nuclear Regulatory Commission (NRC) is investigating the performance of containments subject to severe accidents. This work is being performed by Sandia National Laboratories (SNL). In 1987, a 1:6-scale Reinforced Concrete Containment (RCC) model was tested to failure. The failure mode was a liner tear. As a result, a separate effects test program has been conducted to investigate liner tearing. This paper discusses the design of test specimens and the results of the testing. The post-test examination of the 1:6-scale RCC model revealed that the large tear was not an isolated event. Other small tears in similar locations were also discovered. All tears occurred near the insert-to-liner transition which is also the region of closest stud spacing. Also, all tears propagated vertically, in response to the hoop strain. Finally, all tears were adjacent to a row of studs. The tears point to a mechanism which could involve the liner/insert transition, the liner anchorage, and the material properties. The separate effects tests investigated these effects. The program included the design of three types of specimens with each simulating some features of the 1:6-scale RCC model. The specimens were instrumented using strain gages and photoelastic materials.

The problem of constructing trees given a matrix of interleaf distances is motivated by applications in computational evolutionary biology and linguistics. The general problem is to find an edge-weighted tree which most closely approximates the distance matrix. Although the construction problem is easy when the tree exactly fits the distance matrix, optimization problems under all popular criteria are either known or conjectured to be NP-complete. In this paper we consider the related problem where we are given a partial order on the pairwise distances, and wish to construct (if possible) an edge-weighted tree realizing the partial order. In particular we are interested in partial orders which arise from experiments on triples of species, which determine either a linear ordering of the three pairwise distances (called Total Order Model or TOM experiments) or only the pair(s) of minimum distance apart (called Partial Order Model or POM experiments). The POM and TOM experimental model is inspired by the model proposed by Kannan, Lawler, and Warnow for constructing trees from experiments which determine the rooted topology for any triple of species. We examine issues of construction of trees and consistency of TOM and POM experiments, where the trees may either be weighted or unweighted. Using these experiments to construct unweighted trees without nodes of degree two is motivated by a similar problem studied by Winkler, called the Discrete Metric Realization problem, which he showed to be strongly NP-hard. We have the following results: Determining consistency of a set of TOM or POM experiments is NP-Complete whether the tree is weighted or constrained to be unweighted and without degree two nodes. We can construct unweighted trees without degree two nodes from TOM experiments in optimal O(n{sup 3}) time and from POM experiments in O(n{sup 4}) time.

Semiconductor ring lasers are being developed for use as direct-waveguide-coupled sources for photonic integrated circuits. This report describes the results of our research and development of this new class of diode lasers. We have fabricated and characterized semiconductor ring lasers which operate continuous-wave at room temperature with a single-frequency output of several milliwatts. Our work has led to an increased understanding of the operating behavior of these lasers and to the development of two new types of advanced devices. The interferometric ring diode laser uses a coupled-cavity structure to improve the level of single-frequency performance. And, the unidirectional ring diode laser uses an active crossover waveguide to promote lasing in a single ring direction with up to 96% of the output emitted in the preferred lasing direction.

Results from fundamental investigations of low-temperature plasma systems were used to improve chamber-to-chamber reproducibility and reliability in commercial plasma-etching equipment. The fundamental studies were performed with a GEC RF Reference Cell, a laboratory research system designed to facilitate experimental and theoretical studies of plasma systems. Results and diagnostics from the Reference Cell studies were then applied to analysis and rectification of chamber-to-chamber variability on a commercial, multichamber, plasma reactor. Pertinent results were transferred to industry.

Compression seals are commonly used in electronic components. Because glass has such a low fracture toughness, tensile residual stresses must be kept low to avoid crackS. N. Burchett analyzed a variety of compression pin seals to identify mechanically optimal configurations when work hardened Alloy 52 conductor pins are sealed in a 304 stainless steel housing with a Kimble TM-9 glass insulator. Mechanical property tests on Alloy 52, have shown that the heat treatments encountered in a typical glass sealing cycle are capable of annealing the Alloy 52 pins, increasing ductility and lowering the yield strength. Since most seal analyses are routinely based on unannealed Alloy 52 properties, a limited study has been performed to determine the design impact of lowering the yield strength of the pins in a typical compression seal. Thermal residual stresses were computed in coaxial compression seals with annealed pins and the results then were used to reconstruct design guidelines following the procedures employed by Miller and Burchett. Annealing was found to significantly narrow the optimal design range (as defined by a dimensionless geometric parameter). The Miller-Burchett analyses which were based on very coarse finite element meshes and a 50 ksi yield strength fortuitously predicted an overly conservative design range that is a subset of the narrow design window prevalent when the yield strength is assumed to be 34 ksi. This may not remain true for lower yield strengths. The presence of pin wetting was shown to exacerbate the glass stress state. The time is right to develop a modern and enhanced set of design guidelines which could address new material systems, three dimensional geometries, and viscoelastic effects.

With ever increasing processor and memory speeds, new methods to overcome the ``I/O bottleneck`` need to be found. This is especially true for massively parallel computers that need to store and retrieve large amounts of data fast and reliably, to fully utilize the available processing power. We have designed and implemented a parallel file system, that distributes the work of transferring data to and from mass storage, across several I/O nodes and communication channels. The prototype parallel file system makes use of the existing single threaded file system of the Sandia/University of New Mexico Operating System (SUNMOS). SUNMOS is a joint project between Sandia National Laboratory and the University of New Mexico to create a small and efficient OS for Massively Parallel (MP) Multiple Instruction, Multiple Data (MIMD) machines. We chose file striping to interleave files across sixteen disks. By using source-routing of messages we were able to increase throughput beyond the maximum single channel bandwidth the default routing algorithm of the nCUBE 2 hypercube allows. We describe our implementation, the results of our experiments, and the influence this work has had on the design of the Performance-oriented, User-managed, Messaging Architecture (PUMA) operating system, the successor to SUNMOS.

An automatic phase identification system that employs a neural network approach to classifying seismic event phases is described. Extraction of feature vectors used to distinguish the different classes is explained, and the design and training of the neural networks in the system are detailed. Criteria used to evaluate the performance of the neural network approach are provided.

Satellite electronics may be subjected to a large fluence of protons from the Van Allen belt and from solar flares. To determine if unhardened electronics will survive a radiation environment, the total ionizing dose and displacement damage to the electronics must be determined. Several computer codes are available for modeling proton transport, ranging in complexity for a very-efficient straight-line approximation to general-geometry time-dependent Monte Carlo transport, with corresponding increase in computer run time. For most satellite applications, neutrons can be neglected in the analysis. However, neutrons may be important for modeling heavily shielded compartments for personnel and electronics.

This report describes the activities and results of an LDRD entitled Sensor Based Process Control. This research examined the needs of the plating industry for monitor and control capabilities with particular emphasis on water effluent from rinse baths. A personal computer-based monitor and control development system was used as a test bed.

A large-scale brine inflow test was conducted 655 m below ground surface in a cylindrical test room at the Waste Isolation Pilot Plant (WIPP). This test was the first large-scale WIPP test that allowed periodic access to a sealed, monitored excavation. The test was designed to characterize the environment within the sealed test room (Room Q) and to examine the surrounding host rock to quantify such characteristics as near-surface resistivity and permeability in the formation surrounding the room. Testing began with room boring in July 1989. Data in this report were collected from the time of test start-up through November 25, 1991. Relative humidity, barometric pressure, and temperature were measured in the sealed environment of the test room. Formation closure rates and electrical resistance of the formation close to the room surface were measured to determine the response of the host rock around Room Q. Brine was collected periodically to quantify the amount of inflow from large-scale openings. Results of the measurements are presented in a series of graphs. This report also describes the features of the test

Pore-pressure and fluid-flow tests were performed in 15 boreholes drilled into the bedded evaporites of the Salado Formation from within the Waste Isolation Pilot Plant (WIPP). The tests measured fluid flow and pore pressure within the Salado. The boreholes were drilled into the previously undisturbed host rock around a proposed cylindrical test room, Room Q, located on the west side of the facility about 655 m below ground surface. The boreholes were about 23 m deep and ranged over 27.5 m of stratigraphy. They were completed and instrumented before excavation of Room Q. Tests were conducted in isolated zones at the end of each borehole. Three groups of 5 isolated zones extend above, below, and to the north of Room Q at increasing distances from the room axis. Measurements recorded before, during, and after the mining of the circular test room provided data about borehole closure, pressure, temperature, and brine seepage into the isolated zones. The effects of the circular excavation were recorded. This data report presents the data collected from the borehole test zones between April 25, 1989 and November 25, 1991. The report also describes test development, test equipment, and borehole drilling operations.

The multiplication of a vector by a matrix is the kernel computation of many algorithms in scientific computation. A fast parallel algorithm for this calculation is therefore necessary if one is to make full use of the new generation of parallel supercomputers. This paper presents a high performance, parallel matrix-vector multiplication algorithm that is particularly well suited to hypercube multiprocessors. For an n x n matrix on p processors, the communication cost of this algorithm is O(n/{radical}p + log(p)), independent of the matrix sparsity pattern. The performance of the algorithm is demonstrated by employing it as the kernel in the well-known NAS conjugate gradient benchmark, where a run time of 6.09 seconds was observed. This is the best published performance on this benchmark achieved to date using a massively parallel supercomputer.

Radioactive spent fuel assemblies are a source of hazardous waste that will have to be dealt with in the near future. It is anticipated that the spent fuel assemblies will be transported to disposal sites in spent fuel transportation casks. In order to design a reliable and safe transportation cask, the maximum cladding temperature of the spent fuel rod arrays must be calculated. A comparison between numerical calculations using commercial thermal analysis software packages and experimental data simulating a horizontally oriented spent fuel rod array was performed. Twelve cases were analyzed using air and helium for the fill gas, with three different heat dissipation levels. The numerically predicted temperatures are higher than the experimental data for all levels of heat dissipation with air as the fill gas. The temperature differences are 4{degree}C and 23{degree}C for the low heat dissipation and high heat dissipation, respectively. The temperature predictions using helium as a fill gas are lower for the low and medium heat dissipation levels, but higher at the high heat dissipation. The temperature differences are 1{degree}C and 6{degree}C for the low and medium heat dissipation, respectively. For the high heat dissipation level, the temperature predictions are 16{degree}C higher than the experimental data. Differences between the predicted and experimental temperatures can be attributed to several factors. These factors include experimental uncertainty in the temperature and heat dissipation measurements, actual convection effects not included in the model, and axial heat flow in the experimental data. This work demonstrates that horizontally oriented spent fuel rod surface temperature predictions can be made using existing commercial software packages. This work also shows that end effects will be increasingly important as the amount of dissipated heat increases.

This bulletin discusses the following: decontamination of polluted water by using a photocatalyst to convert ultraviolet energy into electrochemical energy capable of destroying organic waste and removing toxic metals; monitoring oil spills with SAR by collecting data in digital form, processing the data, and creating digital images that are recorded for post-mission viewing and processing; revitalization of a solar industrial process heat system which uses parabolic troughs to heat water for foil production of integrated circuits; and an electronic information system, EnviroTRADE (Environmental Technologies for Remedial Actions Data Exchange) for worldwide exchange of environmental restoration and waste management information.