Many industrial processes are highly non-linear and complex. Their simulation with first-principle or conventional input-output correlation models is not satisfactory, either because the process physics is not well understood, or it is so complex that direct simulation is either not adequately accurate, or it requires excessive computation time, especially for on-line applications. Artificial intelligence techniques (neural networks, expert systems, fuzzy logic) or their combination with simple process-physics models can be effectively used for the simulation of such processes. Feedforward (static) neural networks (FNNs) can be used effectively to model steady-state processes. They have also been used to model dynamic (time-varying) processes by adding to the network input layer input nodes that represent values of input variables at previous time steps. The number of previous time steps is problem dependent and, in general, can be determined after extensive testing. This work demonstrates that for dynamic processes that do not vary fast with respect to the retraining time of the neural network, an adaptive feedforward neural network can be an effective simulator that is free of the complexities introduced by the use of input values at previous time steps.

We have studied the energy resolution of a prototype gas tracking calorimeter in a test beam at Fermilab as part of the detector development program for the MINOS long baseline neutrino oscillation experiment. The calorimeter consisted of 25 layers of 1.5 inch thick steel plates interleaved with planes of aluminum proportional tubes. The tube cells are square, with 0.9 cm edges and open tops. Cathode strips were used for read out transverse to the wire cells. The tubes operated with a nonflammable gas mixture of 88% CO{sub 2}, 9.5% isobutane and 2.5% argon which gave an operating range of >500 V (limited by the electronics). We read out the wire signals on the tubes and in some configurations the cathode stripe as well. We studied positrons, pions and muons over a momentum range of 2.5-30 GeV/c and achieved energy resolutions of about 40%/{radical}E for EM and 71%/{radical}E for hadronic showers.

High-purity tantalum single crystal cylinders oriented with [011] parallel to the cylinder axis were deformed 10, 20, and 30 percent in compression. The engineering stress-strain curve exhibited an up-turn at strains greater than {approximately}20% while the samples took on an ellipsoidal shape during testing, elongated along the [100] direction with almost no dimensional change along [0{bar 1}1]. Two orthogonal planes were selected for characterization using Orientation Imaging Microscopy (OIM): one plane containing [100] and [011] (longitudinal) and the other in the plane containing [0{bar 1}1] and [011] (transverse). OIM revealed patterns of alternating crystal rotations that develop as a function of strain and exhibit evolving length scales. The spacing and magnitude of these alternating misorientations increases in number density and decreases in spacing with increasing strain. Classical crystal plasticity calculations were performed to simulate the effects of compression deformation with and without the presence of friction. The calculated stress-strain response, local lattice reorientations, and specimen shape are compared with experiment.

Pyrochemical conditioning of spent nuclear fuel for the purpose of final disposal is currently being demonstrated at Argonne National Laboratory (ANL), and ongoing research in this area includes the demonstration of this process on spent oxide fuel. In conjunction with this research a pilot scale of the preprocessing stage is being designed by ANL-W to demonstrate the in situ hot cell capability of the chemical reduction stage. An impurity evaluation was completed for a Li/LiCl salt matrix in the presence of spent LWR uranium oxide fuel. A simple analysis was performed in which the sources of impurities in the salt matrix were only from the cell atmosphere. Only reactions with the lithium were considered. The levels of impurities were shown to be highly sensitive system conditions. A predominance diagram for the Li-O-N system was constructed for the device, and the general oxidation, nitridation and combined reactions were calculated as a function of oxygen and nitrogen partial pressure. These calculations and hotcell atmosphere data were used to determine the total number and type of impurities expected in the salt matrix and the mass rate for the device was determined.

Presolar silicon carbide grains form in a variety of types of stars, including asymptotic giant branch red giant stars and supernovae. The dominant mechanisms of heavy element nucleosynthesis, the s-process and r-process, are thought to occur in AGB stars and supernovae, respectively. We have previously reported that mainstream SiC grains have strong enrichments in the s-process isotopes of Sr, Zr and Mo. We report here the first measurements of Mo isotopes in X-type SiC grains, which have previously been identified as having formed from supernova ejecta.

Although a variety of types of pre-solar SiC grains have been classified by their C, N, and Si isotopic composition, the majority of such grains are so-called mainstream grains and are believed to have come from asymptotic giant branch stars [1]. We have previously reported the Mo isotopic compositions of presolar SiC grains whose C, N, and Si isotopic compositions were not known [2]. Since most presolar SiC grains fall in the mainstream group, we assumed that these grains were mainstream. The excellent match of the Mo isotopic data with expectations for nucleosynthesis in AGB stars was consistent with this identification. In order to better understand the distribution of isotopic compositions in presolar grains, we have begun to measure heavy element isotopic compositions of presolar SiC grains of known C, N and Si isotopic composition.

The stability of composite positive and negative electrodes for rechargeable lithium batteries is discussed. Positive electrodes with spinel-type structures that are derived from orthorhombic-LiMnO{sub 2} and layered-MnO{sub 2} are significantly more stable than standard spinel Li[Mn{sub 2}]O{sub 4} electrodes when cycled electrochemically over both the 4-V and 3-V plateaus in lithium cells. Transmission electron microscope data of cycled electrodes have indicated that a composite domain structure accounts for this greater electrochemical stability. The performance of composite Cu{sub x}Sn materials as alternative negative electrodes to amorphous SnO{sub x} electrodes for lithium-ion batteries is discussed in terms of the importance of the concentration of the electrochemically inactive copper component in the electrode.

Air-plasma-sprayed zirconia thermal barrier coatings were subjected to thermal cycling and residual stress evolution in thermally grown oxide scale was studied by micro- and macro-ruby fluorescence spectroscopy. The macro approach reveals that compressive stress in the oxide scale increases with increasing number of thermal cycles (and thus increasing scale thickness), reaching a value of 1.8 GPa at a scale thickness of 3-4 {micro}m (80 cycles). Micro-ruby fluorescence spectroscopy indicates that protrusions of the zirconia top coat into the bond coat act as localized areas of high stress concentration, leading to damage initiation during thermal cycling.

Azimuthal correlations have been studied in heavy ion reactions over a wide range of beam energies. At low incident energies up to 100 MeV/nucleon where collective effects like the directed sidewards flow are generally small, azimuthal correlations provide a useful tool to determine the reaction plane event by event. In the energy regime of the BEVALAC (up to 1 GeV/nucleon for heavy ions) particular emission patterns, i.e. azimuthal correlations of nucleons and light nuclei with respect to the reaction plane, have been associated with the so called squeeze out and sidesplash effects. These effects are of particular interest because of their sensitivity to the equation of state at the high baryon density which is build up during the collision process. Angular distributions similar to the squeeze out have been observed for pions at the SIS in Darmstadt as well as from the EOS - collaboration. Recently also the sideward flow was measured for pions and kaons. However, the origin of the signal in the case of produced mesons is thought to be of a different nature than that for the nucleon flow. At the AGS, azimuthally anisotropic event shapes have been reported from the E877 collaboration for the highest available …

The depolymerization of organosolv-derived lignins by bases in methanol or ethanol solvent was studied in rapidly heated batch microreactors. The conversion of lignin to ether solubles by KOH in methanol or ethanol was rapid at 290 "C, reaching the maximum value within 10-15 minutes. An excess of base relative to Lignin monomer units was required for maximum conversion. Strong bases (KOH, NaOH, CSOH) convert more of the lignin to ether soluble material than do weaker bases LiOH, Ca(OH)2, and NacCO2). Ethanol and methanol are converted to acetic and formic acid respectively under the reaction conditions with an activation energy of approximately 50 kcal/mol. This results in a loss of solvent, but more importantly neutralizes the base catalyst, halting forward progress of the reaction.

The goals are to improve performance and reduce costs; the variables tested are fuel fabrication and assembly tolerances and cladding materials. Significant results are: goal lifetimes achieved; D9/HT9 alloys superior--reduced swelling potential duct mechanical attachment methods viable; test performance per design predictions.

This method provides a structured and cost-oriented way to determine risks associated with loss and destruction of industrial security interests consisting of material assets and human resources. Loss and destruction are assumed to be adversary perpetrated, high-impact events in which the health and safety of people or high-value property is at risk. This concept provides a process for: (1) assessing effectiveness of all integrated protection system, which includes facility operations, safety, emergency and security systems, and (2) a qualitative prioritization scheme to determine the level of consequence relative to cost and subsequent risk. The method allows managers the flexibility to establish asset protection appropriate to programmatic requirements and priorities and to decide if funding is appropriate. The evaluation objectives are to: (1) provide for a systematic, qualitative tabletop process to estimate the potential for an undesirable event and its impact; and (2) identify ineffective protection and cost-effective solutions.

FFTF was originally designed/constructed/operated to develop LMFBR fuels and materials. Inherent safety became a major focus of the US nuclear industry in the mid 1980`s. The inherent safety characteristics of LMFBRs were recognized but additional enhancement was desired. The presentation contents are: Fast Flux Test Facility history and status; Overview of contract activities; Summary of loss of flow without scram with GEMs testing; and Summary of pump start with GEMs testing.

This paper discusses the smart materials/technologies test bed at SRS and the development of an industry-university-laboratory team to support the SRS smart materials and technology demonstration.

Argonne National Laboratory and Lockheed Martin Idaho Technologies Company have jointly participated in the Department of Energy`s (DOE) Waste Isolation Pilot Plant (WIPP) Transuranic Waste Characterization Program since 1990. Intrusive examinations have been conducted in the Waste Characterization Area, located at Argonne-West in Idaho Falls, Idaho, on over 200 drums of mixed contact-handled transuranic waste. This is double the number of drums characterized since the last update at the 1995 Waste Management Conference. These examinations have provided waste characterization information that supports performance assessment of WIPP and that supports Lockheed`s compliance with the Resource Conservation and Recovery Act. Operating philosophies and corresponding regulatory permits have been broadened to provide greater flexibility and capability for waste characterization, such as the provision for minor treatments like absorption, neutralization, stabilization, and amalgamation. This paper provides an update on Argonne`s intrusive characterization permits, procedures, results, and lessons learned. Other DOE sites that must deal with mixed contact-handled transuranic waste have initiated detailed planning for characterization of their own waste. The information presented herein could aid these other storage and generator sites in further development of their characterization efforts.

Pre-and post-test analytical predictions of the dynamic behavior of a 1:10 scale model Reinforced Concrete Containment Vessel are presented. This model, designed and constructed by the Nuclear Power Engineering Corp., was subjected to seismic simulation tests using the high-performance shaking table at the Tadotsu Engineering Laboratory in Japan. A group of tests representing design-level and beyond-design-level ground motions were first conducted to verify design safety margins. These were followed by a series of tests in which progressively larger base motions were applied until structural failure was induced. The analysis was performed by ANATECH Corp. and Sandia National Laboratories for the US Nuclear Regulatory Commission, employing state-of-the-art finite-element software specifically developed for concrete structures. Three-dimensional time-history analyses were performed, first as pre-test blind predictions to evaluate the general capabilities of the analytical methods, and second as post-test validation of the methods and interpretation of the test result. The input data consisted of acceleration time histories for the horizontal, vertical and rotational (rocking) components, as measured by accelerometers mounted on the structure's basemat. The response data consisted of acceleration and displacement records for various points on the structure, as well as time-history records of strain gages mounted on the reinforcement. This paper …

MicroElectricalMechanical Systems (MEMS) were subjected to a vibration environment that had a peak acceleration of 120g and spanned frequencies from 20 to 2000 Hz. The device chosen for this test was a surface-micromachined microengine because it possesses many elements (springs, gears, rubbing surfaces) that may be susceptible to vibration. The microengines were unpowered during the test. The authors observed 2 vibration-related failures and 3 electrical failures out of 22 microengines tested. Surprisingly, the electrical failures also arose in four microengines in the control group indicating that they were not vibration related. Failure analysis revealed that the electrical failures were due to shorting of stationary comb fingers to the ground plane.

Hydrogen embrittlement describes a group of phenomena leading to the degradation of metal alloy properties. The hydrogen concentration in the alloy can be used as an indicator for the onset of embrittlement. A neutron tomography system has been optimized to perform nondestructive detection of hydrogen concentration in titanium aircraft engine compressor blades. Preprocessing of back projection images and postprocessing of tomographic reconstructions are used to achieve hydrogen concentration sensitivity below 200 ppm weight. This paper emphasizes the postprocessing techniques which allow automated reporting of hydrogen concentration.

In this work the basic Finite Element Tearing and Interconnecting (FETI) linear system solver and the PARPACK eigensolver are combined to compute the smallest modes of symmetric generalized eigenvalue problems that arise from structures modeled primarily by solid finite elements. Problems with over one million unknowns are solved. A comprehensive and relatively self-contained description of the FETI method is presented.

Investigation and evaluation of a complex system is often accomplished through the use of performance measures based on system response models. The response models are constructed using computer-generated responses supported where possible by physical test results. The general problem considered is one where resources and system complexity together restrict the number of simulations that can be performed. The levels of input variables used in defining environmental scenarios, initial and boundary conditions and for setting system parameters must be selected in an efficient way. This report describes an algorithmic approach for performing this selection.

In this paper, we examine the logic and framework for the development of a capability to immobilize excess Russian weapons plutonium by the year 2004. The initial activities underway in Russia, summarized here, include engineering feasibility studies of the immobilization of plutonium-containing materials at the Krasnoyarsk and Mayak industrial sites. In addition, research and development (R&D) studies are underway at Russian institutes to develop glass and ceramic forms suitable for the immobilization of plutonium-containing materials, residues, and wastes and for their geologic disposal.

Prediction of the amount of water that may seep into the waste emplacement drifts is an important aspect of assessing the performance of the proposed geologic nuclear waste repository at Yucca Mountain, Nevada. The repository is to be located in thick, partially saturated fractured tuff that will be heated to above-boiling temperatures as a result of heat generation from the decay of nuclear waste. Since water percolating down towards the repository will be subject to vigorous boiling for a significant time period, the superheated rock zone (i.e., rock temperature above the boiling point of water) can form an effective vaporization barrier that reduces the possibility of water arrival at emplacement drifts. In this paper, we analyze the behavior of episodic preferential flow events that penetrate the hot fractured rock, and we evaluate the impact of such flow behavior on the effectiveness of the vaporization barrier.

Sodium/potassium (NaK) liquid metal coolant, contaminated with fission products from the core meltdown of Experimental Breeder Reactor I (EBR-I) and classified as a mixed waste, has been deactivated and converted to a contact-handled, low-level waste at Argonne's Sodium Component Maintenance Shop and land disposed at the Radioactive Waste Management Complex. Treatment of the EBR-I NaK involved converting the sodium and potassium to its respective hydroxide via reaction with air and water, followed by conversion to its respective carbonate via reaction with carbon dioxide. The resultant aqueous carbonate solution was solidified in 55-gallon drums. Challenges in the NaK treatment involved processing a mixed waste which was incompletely characterized and difficult to handle. The NaK was highly radioactive, i.e. up to 4.5 R/hr on contact with the mixed waste drums. In addition, the potential existed for plutonium and toxic characteristic metals to be present in the NaK, resultant from the location of the partial core meltdown of EBR-I in 1955. Moreover, the NaK was susceptible to degradation after more than 40 years of storage in unmonitored conditions. Such degradation raised the possibility of energetic exothermic reactions between the liquid NaK and its crust, which could have consisted of potassium superoxide as well …

The Spallation Neutron Source (SNS) is a second generation pulsed neutron source and is presently in the fourth year of a seven-year construction cycle at Oak Ridge National Laboratory. A collaboration of six national laboratories (ANL, BNL, LANL, LBNL, ORNL, TJNAF) is responsible for the design and construction of the various subsystems. The operation of the facility will begin in 2006 and deliver a 1.0 GeV, 1.4 MW proton beam with pulse length of 650 nanosecond at a repetition rate of 60 Hz, on a liquid mercury target. It consists of an RF volume H{sup -} source of 50 mA peak current at 6% duty; an all electrostatic Low-Energy Beam Transport (LEBT) which also serves as a first stage beam chopper with {+-} 25 ns rise/fall time; a 402.5 MHz, 4-vane Radio-Frequency Quadrupole (RFQ) for acceleration up to 2.5 MeV; a Medium Energy Beam Transport (MEBT) housing a second stage chopper (<{+-} 10ns rise/fall), an adjustable beam halo scraper, and diagnostics devices; a 6-tank Drift Tube Linac (DTL) with permanent magnet quadrupoles up to 87 MeV; an 805 MHz, 4-module, Side Coupled Cavity Linac (CCL) up to 186 MeV; an 805 MHz, superconducting RF (SRF) linac with eleven medium beta …