Spring 2004

May 12, 2004
Dr. Erkin Sidick
Jet Propulsion Laboratory

Title: Silica-Based Planar Lightwave Circuits



May 5, 2004
Sergio Aragon
Professor of Chemistry
Department of Chemistry & Biochemistry
San Francisco State University

Title: Accurate Hydrodynamics of Biomolecules from the Boundary Element Method

In the past 20 years transport properties of macromolecules have been calculated by representing a molecule as a collection of beads. Bead methods produce results that are only accurate to about 15% and cannot predict consistent results for both the translational and rotational diffusion coefficients of molecules. The boundary element method of Youngren-Acrivos has been implemented for the computation of translational and rotational diffusion coefficients of arbitrarily shaped molecules to very high accuracy and precision. A solvent accessible surface is computed from the x-ray or molecular modeled structure and is triangulated. Integration of the Oseen Tensor over triangular surface elements is done essentially exactly. To obtain the best accuracy it is necessary to extrapolate to infinite number of surface elements. Diffusion coefficients can be calculated to 0.1% accuracy, which is 10 times more precise than can be typically measured. We have performed extensive computations on proteins and DNA. We find that a uniform hydration layer of about 1.1 angstroms thick allows us to compute hydrodynamic properties of monomeric proteins in agreement with experiment. This includes the rotational diffusion tensors, the average translational diffusion coefficient, the intrinsic viscosity, the sedimentation coefficient, and the specific volume. For multimeric proteins, we detect a systematic difference between the crystal structure and the structure in solution, a first for hydrodynamic methods.



April 28, 2004
Wilson Ho
Department of Physics and Astronomy and Department of Chemistry
University of California, Irvine

Title: Nanoscience at the Atomic and Molecular Scales

The unique capabilities of the scanning tunneling microscope (STM) for spatially resolved imaging, manipulation, spectroscopy, and chemistry enable new opportunities for control and investigation of chemistry and condensed matter at the atomic and molecular scales. Reactants are manipulated into pre-designed configurations to capture the formation of reaction intermediates and subsequent stimulation of reactions. Detailed motions, conformational changes, and energy exchange in single molecules are induced and monitored, revealing their fundamental properties. Manipulation of individual atoms leads to the formation of chains of metal atoms, which allows the realization of optical emission and magnetic systems from the bottom up, starting from single atoms. These atomic chains enable the observation of particle-in-a-box states as well as the effects of single impurities, systematic variation of the structure, composition, and atomic arrangement in alloys, and chemical sensing of single molecules. The interactions of a pair of chains in a break-junction configuration with a single molecule make it possible to visualize and understand the nature of the contacts between a molecule and the metallic leads, which are pertinent to molecular electronics. This approach reveals a way to synthesize chemical systems with novel bonding and structure. The use of spin polarized electrons from the STM tip enables the investigation of magnetic properties at the atomic scales. In this talk, an overview will be given, highlighting a variety of interesting facets of atomically controlled structures, down to those of single atoms and molecules, and the scientific excitement of nanoscience.



April 21, 2004
Prof. Robin L. B. Selinger,
Department of Physics, Catholic University, Washington DC

Title: How things Bend: Dislocation Patterning and Size Effects in Plastic Deformation

Experiments have shown that the plastic deformation of engineering materials depends in an unexpected way on specimen size, a phenomenon thought to arise from the emergence of a characteristic length scale in the cooperative behavior of dislocations. But what controls that length scale, and what is its physical significance? To explore some of the fundamental mechanisms driving size effects, we perform simulation studies of dislocation dynamics and patterning in two dimensions, using idealized atomistic models. A study of screw dislocations shows shear localization in slip bands; size effects are produced by the emergence of a hard layer near the sample surface, where the yield stress is higher than in the bulk. A study of edge dislocation patterning under bending demonstrates nucleation of tilt boundaries and the onset of grain subdivision, and shows a reverse size effect similar to that seen in nanocrystalline solids. These results suggest new approaches to modeling constitutive behavior of materials at the nano- to microscale.



April 14, 2004
Prof. Troy Carter
Dept. of Physics and Astronomy, UCLA

Title: "Studies of MHD turbulence in a laboratory plasma".

Electromagnetic turbulence is thought to play an important role in plasmas in the laboratory (e.g. transport in magnetic fusion devices) and in astrophysical settings (e.g. interstellar medium, accretion disks). As a specific example, I will discuss the possible role of magnetohydrodynamic (MHD) turbulence in radiatively-inefficient accretion flows. Here it is thought that protons are heated preferentially over electrons through a turbulent cascade from large scale magnetic instabilities (such as the magnetorotational instability) down to dissipative scales. I will use this to motivate a new laboratory experiment studying nonlinear interactions between Alfven waves. In incompressible MHD turbulence, nonlinear interactions between counter-propagating shear Alfven waves are fundamental to the turbulent energy cascade. To study these interactions, large amplitude Alfven waves are created using both a resonant cavity and antennas in the LArge Plasma Device (LAPD) at UCLA. I will discuss the details of these experiments, including reporting on early observations of three-wave interactions between shear waves and a low-frequency compressible fluctuation which is potentially a slow magnetosonic wave.



March 30, 2004 SPECIAL EVENT!
Alan Heeger (received the Nobel Prize for chemistry in 2000).

Click here for details.



March 15, 2004
Dr. Gang Lu
Dept of Physics & Division of Engineering and Applied Science
Harvard University

Title: Multiscale Modeling of Materials Based on Ab initio Calculations

Recent progress has made possible the modeling of material properties across various length and time scales with ab initio calculations. Such approaches are capable to predict materials behavior at macroscopic scales based on microscopic physics that ab initio calculations provide. I will review two ab initio based multiscale approaches that seek to link different length scales sequentially and concurrently, respectively. The first approach relies on a variational continuum formulation of dislocations, in the spirit of the Peierls-Nabarro model. The key ingredient in this approach is the ab initio determined Gamma surface, which can capture the effect of chemistry on mechanical properties of materials. The second approach involves concurrent coupling of electronic degrees of freedom via a real-space orbital-free density functional theory with classic atomic degrees of freedom via the empirical embedded-atom-method, and continuum degrees of freedom via the finite-element based quasicontinuum method. The unified approach that combines quantum mechanics, classic atomistics, and continuum mechanics allows us to study materials behavior at larger length-scales with desirable accuracy and predictive power. The application of these approaches to interesting material problems will be presented. The perspective of computational design of materials based on multiscale simulation strategies will also be discussed.



March 12, 2004
Dr. Guoping Zhang
Department of Physics
Indiana State University

Title: Emerging opportunities for computational materials theory where femtosecond technology and nanoscience intersect

The ever-growing femtosecond (fs) laser technology has been revolutionizing a broad scope of physics community. It has attracted substantial attentions from material scientists as well as industrial researchers and major funding agencies. This technology has a unique capability, not commonly shared by other techniques, to snapshot charge and spin dynamics and to separate it from the dynamics of molecules, liquids and solids. With the sophisticated first-principles and model Hamiltonian calculation, the computational materials theory is in an excellent position to investigate dynamics on a nanometer and fs scale. In this talk, I will show some promising opportunities on the borders between femtosecond technology and nanoscience. I will firstly introduce ultrafast dynamics by examples in a broad scope of the current field. Then I will focus electron and lattice dynamics in a nanosystem: C60. In particular, I will demonstrate why the laser has the capability to selectively probe a few specific vibrational modes, and examine how the electron correlation affects the dynamical response. Finally, I will point out opportunities for computational materials theory in semiconductor, ferromagnets, superconductors and biological systems.



March 10, 2004
Anderson Janotti
Metals and Ceramics Division and Center for Computational Sciences
Oak Ridge National Laboratory
Oak Ridge, TN 37831
USA

Title: "Larger atoms move faster in Metals"

Complicated mass transport processes in solids can be understood on the grounds of fundamental microscopic atomic-level diffusion mechanisms. The kinetics of these thermally activated processes is determined by the migration and formation energies of point defects that mediate the motion of host- or foreign-atoms in solids. Understanding diffusion of substitutional alloying elements is fundamental in designing new and novel grades of transition-metal superalloys for high-temperature applications, such as, jet and rocket propulsion, e.g., turbofans for the Boeing 777 or the Space Shuttle main engines. Unfortunately, traditional practice of alloy design has tended to be rather Edisonian in its approach - a considerable amount of empiricism and trial-and-error testing has been employed to optimise the chemistry of the superalloy grades used. In this presentation I will report theoretical interdiffusion study of 4d and 5d transition metals in Ni base alloy. We observe a rather surprisingly and counterintuitive correlation with the atomic size of the interdiffusion transition metals species: larger atoms display lower diffusion activation energies, e.g., the largest the radius the faster it diffuses. Our results not only explain the recent puzzling experimental observations, we generalize them to the more intriguing 3d transition-metal series.



March 8, 2004
Dr. Vladimir Antropov
Ames Laboratory

Title: Magnetic short range order effects in the itinerant magnet

A theoretical analysis of magnetic order as a function of temperature in 3d magnets is performed using time dependent density functional theory of spin dynamics. A short range order specific for itinerant magnets is revealed. The existence of such order allows us for the first time to resolve several longstanding problems in magnetism theory: the behavior of the high temperature susceptibility and the critical temperature of magnetic phase transition in itinerant magnets. A critical analysis of reliability of the time dependent density functional theory results and comparison with other techniques is provided. Our prediction allows us to modify a traditional interpretation of many experimental results in the magnetism area and leads to the qualitatively new level of analyzing finite temperature properties of magnets.



March 3, 2004
Dr. Erick Draeger

Title: First-principles simulations of silicon quantum dots: the role of synthesis

Silicon quantum dots have been shown to exhibit visible luminescence whose color can be precisely tuned by controlling the size of the cluster, due to the fact that the enhancement of the optical absorption gap is primarily due to quantum confinement effects. These unique optical properties, coupled with their compatibility with existing silicon-based and nano-biological technologies, make these systems very promising for numerous applications. However, significant discrepancies in the optical properties of silicon nanoclusters have been observed. To this end, we studied the effect of synthesis conditions on the structural and optical properties of silicon nanoparticles, using first-principles molecular dynamics and quantum Monte Carlo. Simulations modeling a high temperature synthesis process consistently yield non-crystalline structures with large optical gaps in good agreement with experimental measurements. In addition, we find novel reconstructions of crystalline silicon nanoclusters which are lower in energy and have significantly larger optical absorption gaps than the bulk-derived structures used in previous theoretical studies. This suggests that nanocluster structure can vary systematically with synthesis conditions and may strongly influence the resulting optical properties. These results, along with the effect of surface oxygen atoms on the optical absorption gap, point to a strong dependence of the optical gap on angular strain.



March 1, 2004
Kirill Belashchenko
University of Nebraska, Lincoln

Title: Multiscale physics of magnetization reversal in hard magnets

Magnetization reversal in ferromagnets is probably the best known example of hysteresis, or the history dependence of observed properties. One of the main sources of hysteresis is the interaction of moving domain walls with various defects. The role of microstructure (types of defects and patterns formed by them) is particularly important in hard magnets. The problem involves a number of length scales reflecting the range of basic interactions, the size of defects, and the scales of microstructural patterns. Different length scales call for separate treatment, but they all must be linked together in order to understand the reversal mechanisms and the coercivity. I will explain the ingredients of this problem, using as an illustration our results for polytwinned CoPt-type magnets. I will show how first-principles calculations and atomistic simulations are used to study the interaction of domain walls with defects, how the coercivity emerges as a combined effect of domain wall interaction with the microstructure at different length scales, how the microstructural evolution is studied theoretically, and how it is included in magnetic simulations.



February 25, 2004
Dr. Otto F. Sankey
Arizona State University

Title: Theoretical Aspects of Electron Conduction Through Single Molecule Wires

Single molecules, or molecular arrays, are being considered as components for electronic nano-devices to supplement semiconductor technology. The theoretical concepts concerning coherent electron (tunneling) transport through molecules as applied to molecular electronic systems will be reviewed. Many of the basic concepts are understood, but using these to produce quantitative agreement between theory and experiment is still difficult. We discuss the concepts and describe a simple approach that produces a framework to at least understand the length dependence of the electronic current for some model molecules. We apply these concepts to the simple chemically saturated system of alkane chains, and to the bond-alternating pi-system of carotene molecules sandwiched between gold electrodes. We also describe a theoretical study of a model photochromic molecule which can rapidly switch from an "off" state to an "on" enhanced conductivity state by the application of light.



February 23, 2004
Dr. J. M. Rickman
Department of Materials Science and Engineering
Lehigh University
Bethlehem, PA 18015

Title: Some Applications of Materials Theory and Modeling to Problems in Science and Engineering

Over the past few years materials theory and modeling has become an increasingly important subfield of materials research, with its growing importance a result of the enormous increase in computing power and the maturation of a number of computational methodologies. As such, it encompasses a large variety of techniques applied over a correspondingly disparate range of length and time scales to describe structure and properties of wide classes of materials systems. A reference to materials modeling thus connotes one thing to an investigator using atomistic simulations to study the structure of crystalline defects (e.g., surfaces, dislocations) and something greatly different to someone studying the temporal evolution of systems at the micron scale. In this talk I will illustrate the utility of different computational approaches in addressing some problems of significant current interest, namely: 1.) the plastic deformation of metals, 2.) the kinetics of phase transitions and associated microstructural evolution, and 3.) the development of a real-time, atomic-level simulation package for use in materials design. I will conclude with a discussion of promising strategies for (so-called) multiscale modeling in which the investigator attempts to bridge length and time scales by invoking various physical arguments.



February 18, 2004
Allan Hugh MacDonald

Title: Ferromagnetism and Spin Transport in Semiconductors

Success over the past decade in exploiting the spin-dependent transport properties of ferromagnetic metals in information storage systems, has prompted interest in ferromagnetism and spin-transport in semiconductors. I will discuss progress in creating and understanding new classes of ferromagnetic semiconductors, concentrating on (III,Mn)V materials. These ferromagnets are unusual in that their itinerant quasiparticles are strongly spin-orbit coupled. I will explain the importance of this property for the spin-stiffness and anomalous Hall effect of these materials and address its role in spin torque phenomena.



February 4, 2004 @ 3:45 PM
Debi Prasad Choudhary
NRC research Associate
George C. Marshall Space Flight Center/NASA

Title: Multi-height Vector Magnetography of Solar Active Regions

The magnetic field of the sun is responsible for the existance of a hot corona and most of its visible dynamic features, including the explosive events that can effect the near earth space weather. The magnetic field is generated below the visible layer (the photosphere) and erupt into the solar atmosphere. The cross-section of the erupting field structure at the photosphere is observed as an active region (sunspots), above which there exists a complex three-dimensional magnetic "dome". The photospheric vector magnetic field of solar active regions have been measured on a synoptic basic since last 30 years. Yet, the field in chromosphere and corona are largely derived by extrapolating these observations using numerical modeling. In the last few years several exploratory measurements of magnetic fields in spectral lines originating at chromospheric and coronal heights have shown promising results. Recent technological advances to make efficient filters, detectors and polarizing optics makes it possible to design and fabricate the instruments that can be used to measure simultaneously the vector magnetic fields in several layers above the photosphere. In this talk, I shall present a brief review of multi-height magnetography and its importance for solving some of the outstanding question of solar physics.