|
Technologies for the 21st Century
HECC Supplement |
 
|
|
|
|
|||
 |
|
|---|---|
|
|
|
|
|
Volume rendering techniques
|
Volume rendering includes several techniques for visualizing 3D scalar fields.
For example, "isocontouring" extracts constant valued surfaces from
these fields. An alternative is to render directly the 3D data using forward
projection or backward projection methods. Forward projection methods project
samples of the 3D field to the screen, generally using traditional graphics
techniques. Backward projection, commonly referred to as volume raycasting
methods, determines the color of each pixel by finding the subset of the 3D field
that project to the pixel being colored, and then combining them. NSF-supported
research in volume rendering is being conducted at Purdue University.
|
|
|
|
|
Light nuclei studies
|
The properties of light nuclei (up to 40 neutrons and protons) are computed
employing realistic two- and three-nucleon interactions (such as those
illustrated here). Many-body methods are used to compute the properties of a
nucleus for complicated forces that are strongly dependent on the spins and
charge states (isospin) of the nucleons. Unlike the Coulomb force used in atomic
or condensed-matter calculations, there is no useful fundamental theory that
defines this force. One can partially constrain the two-body force by fitting
nucleon-nucleon scattering data, but many-body calculations are required to test
other properties of this force as well as the three-body interaction.
DOE-supported researchers are refining their knowledge of the forces and using
that knowledge to make predictions about the behavior of nuclei.
|
|
|
|
|
Numerical Tokamak
|
A major theme of the DOE-supported Numerical Tokamak Turbulence Project's (NTTP)
gyrofluid research in the past year has been the study of plasma turbulence
suppression methods discovered recently in the Tokamak Fusion Test Reactor (TFTR)
at the Princeton Plasma Physics Laboratory, the DIII-D in Georgia, and the
Japanese JT-60 tokamaks. If these methods scale to larger devices, they could
lead to more attractive and economical fusion reactor designs. There have been
several NTTP studies focusing on the importance of shear in the background flow
in stretching and tearing turbulent eddies, resulting in a suppression of the
turbulence and a reduction in the concomitant transport. The stabilizing effects
of velocity shear have been seen in numerous tokamak experiments. Shaping the
magnetic field in the tokamak experiments introduces physics that is related to
the sheared-flow stabilization method seen in simulations, and has led to
improved confinement in recent experiments. The goal of this multidisciplinary
effort is the realistic simulation of the tokamak plasma turbulence needed to
optimize performance of fusion devices.
|
|
|
|
|
Nonlinear electron
|
NSF-supported researchers have developed a new quantum dynamics simulation scheme
to study highly nonlinear, far-from-equilibrium electron dynamics in nanodevices.
The scheme incorporates the electron-phonon interaction in the mean-field
approximation and dissipation through the Langevin equation. This approach has
been used to study nonlinear electron transport in numerous areas, including
electron mobility in amorphous Silicon.
The figure to the right represents electron wave functions in a double quantum dot: (a) no coupling; (b) electron coupled to phonons at 300 K. Brightness represents the intensity of the electron wave function, and color represents the phase. The incident electron energy is at the resonant transmission peak corresponding to the antisymmetric quasi-bound level. Without the electron-phonon interaction, the probabilities build up and decay equally in the two dots. With the electron-phonon interaction, the probability density first localizes in one dot and then starts to oscillate between the two. |
|
|
|
|
Design optimization techniques
|
Sensitivity analysis is used in design optimization, such as the design of an
aircraft wing that involves integration of several different programs. Optimized
designs can be found automatically by computing sensitivities of each code with
respect to design parameters and applying a gradient-based optimization
technique.
The Automatic Differentiation Tool for ANSI-C (ADIC) developed by researchers in the Mathematics and Computer Science Division at Argonne National Laboratory has been applied to the Coordinate and Sensitivity Calculator for Multidisciplinary Design Optimization (CSCMDO) code developed at NASA's Langley Research center in southern Virginia. CSCMDO fits into the design optimization environment as a means for automatically modifying structured volume grids used in computational fluid dynamics. The ADIC-enhanced version of CSCMDO automatically produces the required volume grid sensitivity. CSCMDO provides a rapid and highly automated 3-D volume grid generation capability that produces changes in the surface and volumetric grids to reflect the perturbations of the baseline system. Algebraic techniques are used to generate and modify block face and volume grids to reflect geometric changes resulting from design optimization. These figures, obtained from an Automatically Differentiated (AD) version of CSCMDO, represent volume grid sensitivity derivatives with respect to different design parameters. |
|
|
|
|
Catalytic properties of MDH
|
DOE-supported researchers have conducted a computational simulation and analysis
of the reaction mechanism of the enzyme malate dehydrogenase (MDH). Encouraged by
preliminary results, researchers calculated the minimum energy surface and
reaction pathway for the interconversion of malate and oxaloacetate catalyzed by
MDH. Analysis of the energy profile shows that solvent effects due to the protein
matrix dramatically alter the intrinsic reactivity of the functional groups
involved in the MDH reactions. The enzyme effectively changes the reaction from
an exothermic reaction in the gas phase to a nearly isoenergetic one in the
protein-solvent environment of MDH. Energy decomposition analysis indicates that
specific MDH residues in the vicinity of the substrate make significant energy
contributions to the stabilization of proton transfer and destabilization of
hydride transfer. This data suggests that amino acids play an important role in
the catalytic properties of MDH, consistent with site-directed mutagenesis
experiments.
|
|
|
|
|
Scientific visualization
|
Cooperation between the EPA Scientific Visualization center and the Space Science
and Engineering center (SSEC) of the University of Wisconsin at Madison has given
EPA scientists 3-D visualization capability using desktop workstations. This
capability joins SSEC's Vis5D system for visualizing the output of atmospheric
and ocean models with composite images from the Geostationary Operational
Environmental Satellite (GOES). This image compares remotely sensed cloud data
from GOES seen as white clouds at the bottom of the image with 3-D cloud data
predicted by the National center for Atmospheric Research's (NCAR) Mesoscale
Meteorological model. Cloud water is seen in blue, rain water in white, and
temperature of the clouds ranges from cold blue to warmer red. Vis5D makes this
interactive exploration possible by compressing data sets so they fit in
workstation memories.
|
|
|
|
|
Biomolecular computing and AChE simulations
|
Biomolecular computing using high performance computing involves extensive, often
complex calculations. This study involves the computer simulation of the enzyme
acetylcholinesterase (AChE),which is responsible for degrading the
neurotransmitter acetylcholine in species from man on down to insects. Due to its
ubiquitous presence in nature and key role in biological systems, AChE is a
target for many commonly used drugs and toxins. Clinical studies supported by NIH
and NSF suggest that acetylcholinesterase inhibitors such as tacrine
(tetrahydro-9-aminoacridine) may be useful in enhancing memory in patients with
Alzheimer's disease. The figure to the right illustrates an AChE-THA complex
structure after 100 picoseconds of molecular dynamics simulation. The ribbon
traces the amino acid chain forming the molecule. Two tacrine molecules are
visible inside the protein.
|
|
|
|
|
Macromolecular Structure
|
Researchers at the UCSD Computational center for Macromolecular Structure (CCMS)
have developed filters and modules for the AVS software package to project the
results from simulations onto molecular surfaces, here showing the hydrophilicity
between the HIV enzyme protease (lower surface) and a bound inhibitor (floating
"balloon"). CCMS develops software for analyzing those structural
features of molecules that play key roles in drug design, such as docking,
electrostatics, and hydrophilicity. CCMS is a joint project of UCSD, The Scripps
Research Institute, and the San Diego Supercomputer center, supported by NSF.
|
|
|
|
|
Molecular dynamics |
Global optimization techniques are central solving macromolecular modeling and
simulation problems, since many fundamental problems in these areas are
formulated as global optimization problems. One aim of this R&D is to develop
a high performance environment on the IBM SP at DOE's Argonne National Laboratory
to support large scale global optimization algorithms and software for solving
global optimization problems arising in the modeling and simulating of large
molecular systems. This figure illustrates applications in protein conformation
and modeling, ionic system configuration, and molecular cluster simulation.
Researchers are using optimization methods to find stable configurations of ionic
systems. In this figure, the stable configuration for an ionic system has the
lowest energy, and therefore can be found by minimizing the energy function for
the system over the configuration space. Stable configurations for a set of small
systems (fewer than 100 ions) have been obtained by using global continuation
algorithms on the IBM SP. The optimal structure with 60 ions is shown here. A
goal of this work is to find the stable configurations for very large systems,
say, systems of 200,000 ions, from which a phase transition of the ionic system
can be observed.
Researchers have applied global continuation algorithms to a set of
Lennard-Jones-potential-based microcluster conformation problems. In this figure,
the optimal structure is shown for small clusters (fewer than 75 atoms) as the
global minimizers of the Lennard-Jones potential functions. This work is being
extended to the general area of molecular cluster simulation, such as simulation
of metal clusters with semi-empirical potentials.
|
|
|
|
|
|
|
