The majority of polyphosphazene material research has concentrated on the linear polymer configuration. However, this represents only one of three potential backbone configurations for phosphazenes. Linear polymers are formed either directly from phosphorus and nitrogen containing precursors or from the ring opening polymerization of hexachlorocyclotriphosphazene. Two other backbone structures can be formed from hexachlorocyclotriphosphazene cyclolinear and cyclomatrix. Cyclolinear are the least studied due to synthetic difficulty. Cyclomatrix polymers represent a more facile method for forming non-linear phosphazenes.

Detailed chemical kinetic reaction mechanisms are developed for all nine chemical isomers of heptane (C{sub 7}H{sub 16}), following techniques and models developed previously for other smaller alkane hydrocarbon species. These reaction mechanisms are tested at high temperatures by computing shock tube ignition delay times and at lower temperatures by simulating ignition in a rapid compression machine. Although the corresponding experiments have not been reported in the literature for most of these isomers of heptane, intercomparisons between the computed results for these isomers and comparisons with available experimental results for other alkane fuels are used to validate the reaction mechanisms as much as possible. Differences in the overall reaction rates of these fuels are discussed in terms of differences in their molecular structure and the resulting variations in rates of important elementary reactions. Reaction mechanisms in this study are works in progress and the results reported here are subject to change, based on model improvements and corrections of errors not yet discovered.

The authors describe the SXMPVO algorithm for performing a visibility ordering zoo-meshed polyhedra. The algorithm runs in practice in linear time and the visibility ordering which it produces is exact.

The Brookhaven Relativistic Heavy Ion Collider (RHIC) provides the unique opportunity to collide polarized proton beams at a center-of-mass energy of up to 500 GeV and luminosities of up to 2 x 10{sup 32} cm{sup {minus}2} s{sup {minus}1}. Such high luminosity and high energy polarized proton collisions will open up the possibility of studying spin effects in hard processes. However, the acceleration of polarized beams in circular accelerators is complicated by the numerous depolarizing spin resonances. Using a partial Siberian snake and a rf dipole that ensure stable adiabatic spin motion during acceleration has made it possible to accelerate polarized protons to 25 GeV at the Brookhaven AGS. After successful operation of RHIC with gold beams polarized protons from the AGS have been successfully injected into RHIC and accelerated using a full Siberian snakes built from four superconducting helical dipoles. A new high energy proton polarimeter was also successfully commissioned. Operation with two snakes per RHIC ring is planned for next year.

This paper discusses the authors efforts to improve the performance of the high-accuracy (HIAC) volume rendering system, based on cell projection, which is used to display unstructured, scientific data sets for analysis. The parallelization of HIAC, using the pthreads and MPI API's, resulted in significant speedup, but interactive frame rates are not yet attainable for very large data sets.

The structure, study topics, straw-man muon collider parameter sets and technical challenges for ''Six-Month Study on High Energy Muon Colliders: Oct'00-Apr'0l'' have been summarized at one month from completion of the study. The extremely high constituent particle energies and luminosities of the parameter sets presented in table 1 continue to suggest that muon colliders could play a central role in exploring and extending the HEP energy frontier. The study has already resulted in encouraging progress in areas such as the final focus lattice design and cost-efficient acceleration.

This reports summarizes the results of postirradiation examinations of a series of unrestrained dilute uranium alloy specimens irradiated to exposures up to 13,000 MWD/T in NaK-containing stainless steel capsules.

Numerical simulations are used to visualize the mixing and combustion induced by explosions of spherical and cylindrical TNT charges. Evolution of the exothermic energy is controlled by mixing (vorticity), which is strongly influenced by wave reflections from confining walls.

This paper describes four general categories of infrastructure interdependencies (physical, cyber, geographic, and logical) as they apply to the water/wastewater infrastructure, and provides an overview of one of the analytic approaches and tools used by Argonne National Laboratory to evaluate interdependencies. Also discussed are the dimensions of infrastructure interdependency that create spatial, temporal, and system representation complexities that make analyzing the water/wastewater infrastructure particularly challenging. An analytical model developed to incorporate the impacts of interdependencies on infrastructure repair times is briefly addressed.

The knowledge gained from scientific observation, experiment, and simulation is linked to the ability to analyze, understand, and manage the generated results. These abilities are increasingly at odds with the current, and future, capabilities to generate enormous quantities of raw scientific and engineering data from instruments, sensors, and computers. Many researchers are currently engaged in activities that seek to create new and novel methods for analyzing, understanding, and managing these vast collections of data. In this work, we present some of our research in addressing a particular type of problem in this broad undertaking. Much the scientific data of interest is in the form of observed, measured, or computed multivariate or multi-component vector field data--with either as physical or color data values. We are currently researching methods and techniques for working with this type of vector data through the use of a novel analysis technique. Our basic approach is to work with the vector field data in its natural physical or color space. When the data is viewed as a functional mapping of a domain, usually an index space, to a range, the physical or color values, potentially interesting characteristics of the data present themselves. These characteristics are useful in …

The authors present an overview of the technical objectives of the Terascale Simulation Tools and Technologies center. The primary goal of this multi-institution collaboration is to develop technologies that enable application scientists to easily use multiple mesh and discretization strategies within a single simulation on terascale computers. The discussion focuses on the efforts to create interoperable mesh generation tools, high-order discretization techniques, and adaptive meshing strategies.

Fragmentation reactions offer a useful tool to study the spectroscopy of halo nuclei, but the large extent of the halo wave function makes the reaction theory more difficult. The simple reaction models based on the eikonal approximation for the nuclear interaction or first-order perturbation theory for the Coulomb interaction have systematic errors that they investigate here, comparing to the predictions of complete dynamical calculations. They find that stripping probabilities are underpredicted by the eikonal model, leading to extracted spectroscopy strengths that are two large. In contrast, the Coulomb excitation is overpredicted by the simple theory. They attribute this to a screening effect, as is well known in the Barkas effect on stopping powers. The errors decrease with beam energy as E{sub beam}{sup -1}, and are not significant at beam energies above 50 MeV/u. At lower beam energies, the effects should be taken into account when extracting quantitative spectroscopic strengths.

None

The availability of high brilliance synchrotron sources, coupled with recent progress in achromatic focusing optics and large area 2D detector technology, have allowed us to develop an X-ray synchrotron technique capable of mapping orientation and strain/stress in polycrystalline thin films with submicron spatial resolution. To demonstrate the capabilities of this instrument, we have employed it to study the microstructure of aluminum thin film structures at the granular and subgranular level. Owing to the relatively low absorption of X-rays in materials, this technique can be used to study passivated samples, an important advantage over most electron probes given the very different mechanical behavior of buried and unpassivated materials.

Dramatic improvement in the efficiency of inorganic and organic light emitting diodes (LEDs and OLEDs) within the last decade has made these devices viable future energy efficient replacements for current light sources. However, both technologies must overcome major technical barriers, requiring significant advances in material science, before this goal can be achieved. Attention will be given to each technology associated with the following major areas of material research: (1) material synthesis, (2) process development, (3) device and defect physics, and (4) packaging. The discussion on material synthesis will emphasize the need for further development of component materials, including substrates and electrodes, necessary for improving device performance. The process technology associated with the LEDs and OLEDs is very different, but in both cases it is one factor limiting device performance. Improvements in process control and methodology are expected to lead to additional benefits of higher yield, greater reliability and lower costs. Since reliability and performance are critical to these devices, an understanding of the basic physics of the devices and device failure mechanisms is necessary to effectively improve the product. The discussion will highlight some of the more basic material science problems remaining to be solved. In addition, consideration will be …

Poor mesh quality is known to adversely affect both solution efficiency and accuracy. There has been considerable research on a wide variety of mesh improvement algorithms, but the impact of these algorithms on applications has been limited because they are typically embedded in particular meshing software packages. To rectify this situation, they are developing a stand-alone mesh quality improvement toolkit called MESQUITE. In this paper, the authors describe the motivation, goals, and design of MESQUITE and give some computational results using the underlying algorithms that show the benefit of such a package.

A model-based detection scheme is developed to passively monitor an ocean acoustic environment along with its associated variations. The technique employs an embedded model-based processor and a reference model in a sequential likelihood detection scheme. The monitor is therefore called a sequential reference detector. The underlying theory for the design is developed and discussed in detail.

In contrast to failure approaches at the lamina level or the micromechanics level the present work concerns failure characterization at the laminate level. Specifically, attention is given to the ultimate failure characterization for quasi-isotropic laminates. This is in further contrast to the commonly used approaches for initial damage or progressive damage. It is shown that the analytical failure forms decompose into two modes, one for out of plane, delamination type failure and one for in plane, fiber controlled type failure. The work here is mainly given over to the delamination mode of failure. Experimental results are presented for laminates in this mode of failure. These results are then integrated with the analytical forms to give a simple criterion for delamination failure.

I compute the v/c correction to the gravitational time delayfor light passing by a massive object moving with speed v, and I finddisagreement with previously published results. It is also argued thatthe speed of gravity formula that was recently used in the conjunction ofJupiter and quasar J0842+1845 is frame dependent.

The science of complexity has moved to center stage within the past few decades. Complex systems range from glasses to the immune system and the brain. Glasses are too simple to possess all aspects of complexity; brains are too complex to expose common concepts and laws of complexity. Proteins, however, are systems where many concepts and laws of complexity can be explored experimentally, theoretically, and computationally. Such studies have elucidated crucial aspects. The energy landscape has emerged as one central concept; it describes the free energy of a system as a function of temperature and the coordinates of all relevant atoms. A second concept is that of fluctuations. Without fluctuations, proteins would be dead and life impossible. A third concept is slaving. Proteins are not isolated systems; they are embedded in cells and membranes. Slaving arises when the fluctuations in the surroundings of a protein dominate many of the motions of the protein proper.

Audio information stored in the undulations of grooves in a medium such as a phonograph record may be reconstructed, with no or minimal contact, by measuring the groove shape using precision metrology methods and digital image processing. The effects of damage, wear, and contamination may be compensated, in many cases, through image processing and analysis methods. The speed and data handling capacity of available computing hardware make this approach practical. Various aspects of this approach are discussed. A feasibility test is reported which used a general purpose optical metrology system to study a 50 year old 78 r.p.m. phonograph record. Comparisons are presented with stylus playback of the record and with a digitally re-mastered version of the original magnetic recording. A more extensive implementation of this approach, with dedicated hardware and software, is considered.

Many turbulent premixed flames of practical interest are statistically stationary. They occur in combustors that have anchoring mechanisms to prevent blow-off and flashback. The stabilization devices often introduce a level of geometric complexity that is prohibitive for detailed computational studies of turbulent flame dynamics. As a result, typical detailed simulations are performed in simplified model configurations such as decaying isotropic turbulence or inflowing turbulence. In these configurations, the turbulence seen by the flame either decays or, in the latter case, increases as the flame accelerates toward the turbulent inflow. This limits the duration of the eddy evolutions experienced by the flame at a given level of turbulent intensity, so that statistically valid observations cannot be made. In this paper, we apply a feedback control to computationally stabilize an otherwise unstable turbulent premixed flame in two dimensions. For the simulations, we specify turbulent in flow conditions and dynamically adjust the integrated fueling rate to control the mean location of the flame in the domain. We outline the numerical procedure, and illustrate the behavior of the control algorithm. We use the simulations to study the propagation and the local chemical variability of turbulent flame chemistry.

The usual approach to deal with noise present in many real-world optimization problems is to take an arbitrary number of samples of the objective function and use the sample average as an estimate of the true objective value. The number of samples is typically chosen arbitrarily and remains constant for the entire optimization process. This paper studies an adaptive sampling technique that varies the number of samples based on the uncertainty of deciding between two individuals. Experiments demonstrate the effect of adaptive sampling on the final solution quality reached by a genetic algorithm and the computational cost required to find the solution. The results suggest that the adaptive technique can effectively eliminate the need to set the sample size a priori, but in many cases it requires high computational costs.

QR and LU factorizations for dense matrices are important linear algebra computations that are widely used in scientific applications. To efficiently perform these computations on modern computers, the factorization algorithms need to be blocked when operating on large matrices to effectively exploit the deep cache hierarchy prevalent in today's computer memory systems. Because both QR (based on Householder transformations) and LU factorization algorithms contain complex loop structures, few compilers can fully automate the blocking of these algorithms. Though linear algebra libraries such as LAPACK provides manually blocked implementations of these algorithms, by automatically generating blocked versions of the computations, more benefit can be gained such as automatic adaptation of different blocking strategies. This paper demonstrates how to apply an aggressive loop transformation technique, dependence hoisting, to produce efficient blockings for both QR and LU with partial pivoting. We present different blocking strategies that can be generated by our optimizer and compare the performance of auto-blocked versions with manually tuned versions in LAPACK, both using reference BLAS, ATLAS BLAS and native BLAS specially tuned for the underlying machine architectures.