Bulletin of the American Physical Society
2007 APS March Meeting
Volume 52, Number 1
Monday–Friday, March 5–9, 2007; Denver, Colorado
Session B27: Focus Session: Computational Nanoscience II-Methods and Applications |
Hide Abstracts |
Sponsoring Units: DMP DCOMP Chair: Sanjay Khare, University of Toledo Room: Colorado Convention Center 301 |
Monday, March 5, 2007 11:15AM - 11:27AM |
B27.00001: Kinetic Monte Carlo simulations of Ag(111) island coarsening Giridhar Nandipati, Yunsic Shim, Jacques Amar, Altaf Karim, Abdelkader Kara, Talat Rahman The results of parallel kinetic Monte Carlo simulations of submonolayer island coarsening on the Ag(111) surface are presented. Our simulations are carried out using a large database of activation barriers which has been generated from previous self-learning kinetic Monte Carlo simulations of small and medium-size clusters. In this database, which includes both single-atom and multi-atom concerted moves, interactions between a central atom and all other adatoms within the first two nearest-neighbor rings are taken into account, while the symmetry of the (111) surface is also used. In order to reach extended time and length-scales we have implemented a novel parallel kinetic Monte Carlo scheme in which processor domains are dynamically assigned in order to minimize boundary events. Preliminary results using an open database corresponding to a true self-learning kinetic Monte Carlo simulation will also be presented. [Preview Abstract] |
Monday, March 5, 2007 11:27AM - 11:39AM |
B27.00002: Self-Learning Off-Lattice Kinetic Monte Carlo method as applied to growth on metal surfaces Oleg Trushin, Abdelkader Kara, Talat Rahman We propose a new development in the Self-Learning Kinetic Monte Carlo (SLKMC) method with the goal of improving the accuracy with which atomic mechanisms controlling diffusive processes on metal surfaces may be identified. This is important for diffusion of small clusters (2 - 20 atoms) in which atoms may occupy Off-Lattice positions. Such a procedure is also necessary for consideration of heteroepitaxial growth. The new technique combines an earlier version of SLKMC [1] with the inclusion of off-lattice occupancy. This allows us to include arbitrary positions of adatoms in the modeling and makes the simulations more realistic and reliable. We have tested this new approach for the case of the diffusion of small 2D Cu clusters diffusion on Cu(111) and found good performance and satisfactory agreement with results obtained from previous version of SLKMC. The new method also helped reveal a novel atomic mechanism contributing to cluster migration.~We have also applied this method to study the diffusion of Cu clusters on Ag(111), and find that Cu atoms generally prefer to occupy off-lattice sites. [1] O. Trushin, A. Kara, A. Karim, T.S. Rahman~ Phys. Rev B 2005 [Preview Abstract] |
Monday, March 5, 2007 11:39AM - 11:51AM |
B27.00003: Diffusion Limited Processes Using Accelerated Molecular Dynamics Erdi Bleda, Xing Gao, Murray Daw We present a systematic microscopic approach to diffusion-limited processes for intermetallic alloys using Accelerated Molecular Dynamics. On-the-fly kinetic Monte Carlo is combined with the Dimer Method to find the saddlepoints exiting a valley, based on energetics from the Embedded Atom Method. With this technique, we compute the tracer diffusivities as a function of composition and temperature for strongly ordered (Cu$_{3}$Au), weakly ordered (Ag-Au) and weakly clustered (Cu-Ni) alloys. [Preview Abstract] |
Monday, March 5, 2007 11:51AM - 12:03PM |
B27.00004: Extension of Mean-Field Nucleation Theory with Long-Range Interactions John A. Venables, James Degraffenreid, Ramon Grima Mean-field nucleation theory is an important tool in understanding various adsorbate-substrate systems, particularly in the context of epitaxial growth. Conventional mean-field theory does not take into account nonlocal interactions, but these can be important in the nucleation and growth of various nanostructures. An approach due to Ovesson [1] is based on the assumption that the change of saddle-point energy in a potential field equals the average changes at the neighboring binding sites, but this assumption is not generally satisfied. We reformulate the theory in a more general sense, as an extension of the work of Grima and Newman [2] and Venables \textit{et al.} [3]. This leads to a continuum mean-field description in a general potential field, in which the transport coefficients are intrinsically connected with the interaction potential and with microscopic parameters. Computational examples are presented for Ge/Si(001) material parameters. \newline \newline [1] S. Ovesson, \textit{PRL} \textbf{88}, 116102 (2002);\newline [2] R. Grima and T.J. Newman, \textit{PRE} \textbf{70}, 036703 (2004);\newline [3] J.A. Venables \textit{et al.}, \textit{PRB} \textbf{74}, 075412 (2006) [Preview Abstract] |
Monday, March 5, 2007 12:03PM - 12:15PM |
B27.00005: Capture-Zone Areas \& the Wigner Distribution: New Case of Universal Scaling of Spacings in Fluctuating Systems A. Pimpinelli, T.L. Einstein When investigating scaling of island sizes during growth in $d$ dimensions, one should consider the distribution of the areas of proximity cells around nucleation centers, i.e. capture zones (CZ). Using data from kinematic Monte Carlo studies,\footnote{ Mulheran et al., PRB {\bf 53} ('96) 10261, {\bf 54} ('96) 11681; EPL {\bf 49} ('00) 617, {\bf 65} (’04) 379. Amar, Family, et al., PRL {\bf 74} ('95) 2066; PRB {\bf 64} (’01) 205404. Evans, Bartelt, et al. PRB {\bf 66} (’02) 235410; SSR {\bf 61} ('06) 1.} we find that the CZ distributions in both $d$ = 1 and $d$ = 2 are well described by the generalized Wigner distribution (GWD) from random-matrix theory: $P_\varrho(s)=as^\varrho\exp(-bs^2)$. $P_\varrho(s)$ accounts for a broad range of fluctuation phenomena, inc.\ the terrace-width distribution (TWD) on vicinal surfaces. For CZ distributions, we find $\varrho = i + d/2$, where $i$ is the critical nucleus size. We present a phenomenological justification by constructing a Langevin equation similar to that used in accounting for the equilibration of TWDs.\footnote{A. Pimpinelli, H. Gebremariam, \& T.L. Einstein, PRL 95 ('05) 246101} We discuss implications for processing and analysis of experimental data. [Preview Abstract] |
Monday, March 5, 2007 12:15PM - 12:27PM |
B27.00006: Feature detection for large-scale molecular dynamics simulations Hyoungki Park, David Richie, Jeongnim Kim, Joseph Gorse, John Wilkins Advances in computer hardware and numerical methods compound the analysis of complex, large-scale evolutionary phenomena. Progress comes from just-in-time analysis and data compression. Real-time multiresolution analysis (RTMRA) on dynamical quantities (e.g., positions and local energies of atoms) -- based on simple Haar wavelets -- compresses data more than 100-fold while retaining 0.1 \AA r.m.s. resolution. Further, RTMRA techniques enable a sophisticated event detection scheme capable of identifying meta-stable structures and detecting infrequent events, e.g., structural transitions, in the presence of thermal noise. As an example, the dynamics over a broad temperature range of silicon defect systems yields visually clear diffusion mechanisms for small silicon interstitial clusters (single-, di-, and tri-interstitial), and initiating growth of extended defects such as the extended {311} defects. [Preview Abstract] |
Monday, March 5, 2007 12:27PM - 12:39PM |
B27.00007: Objective Molecular Dynamics Traian Dumitrica, Richard James We present a generalization of periodic molecular dynamics that we term {\it objective molecular dynamics}. It is a method of doing molecular dynamics for a restricted set of atoms, nonperiodically mapping the time-dependent displacements of this small set of atoms onto the full, typically infinite structure, such that the full structure satisfies exactly the full, unconstrained set of equations of molecular dynamics subject to certain group-invariant initial conditions. The method is applicable to a wide variety of interesting molecular structures including the tails, capsids and other parts of many viruses, carbon nanotubes, many of the common proteins, C$_{60}$ and many other nanostructures now being synthesized, especially via the process of self-assembly. Overall, the strength of the proposed symmetry-based approach is that (i) it heavily reduces the computational effort through a drastic reduction in the number of atoms to be accounted for, (ii) it is compatible with full quantum mechanics, and (iii) the implementation can be done in a general framework, allowing for simulations of a larger class of structures. In addition (iv) the scheme is ideal for obtaining nanomechanical responses since it allows for applying various mechanical deformations. The method is illustrated by simulations of carbon nanotubes. [Preview Abstract] |
Monday, March 5, 2007 12:39PM - 12:51PM |
B27.00008: Decorrelation of samples in Quantum Monte Carlo calculations and applications to metallic nanoclusters Daniel Nissenbaum, Bernardo Barbiellini, Arun Bansil We discuss decorrelation of samples in Quantum Monte Carlo (QMC) ground-state energy calculations for large lithium and water nanoclusters and show how accurate results can be obtained without the need for decorrelating samples. The scaling of the integrated autocorrelation time $\tau$ is analyzed as a function of nanocluster size. $\tau$ is found to scale quadratically in Li nanoclusters, which adds a quadratic factor to the scaling of the total computation time in this metallic case, a factor which does not appear in computations of non-metallic H$_{2}$O nanoclusters. We choose nanoclusters which are relatively large in the context of QMC to demonstrate the application of these techniques - lithium nanoclusters with up to 64 atoms and water nanoclusters with up to 20 molecules. [Preview Abstract] |
Monday, March 5, 2007 12:51PM - 1:03PM |
B27.00009: Linear Scaling NanoScience Simulations for Petascale Computing Zhengji Zhao, Lin-Wang Wang, Juan Meza There are many large-scale nanoscience problems that require \emph {ab initio} accuracy total energy calculations and atomic relaxations. Unfortunately, the traditional direct \emph {ab initio} method scales as $O(N^3)$, where $N$ is the number of atoms in the system, and most of the $O(N)$ methods studied in the last decade have various numerical convergence problems and computer parallelization issues. In this talk, we present an alternative $O(N)$ method which divides the whole system into small fragments. By combining the fragments in an ingenious pattern, the artificial boundary effects of the spatial division can be canceled out. As a reasult, the difference between this method and the direct \emph {ab initio} calculation is smaller than errors introduced by other numerical approximations, and the method scales almost linearly to the number of processors. We have used this method to calculate nanostructures with more than ten thousand atoms using thousands of processors under the conventional planewave pseudopotential approach. We will demonstrate that this approach provides a practical way for future petascale computation in materials/nanomaterials science. [Preview Abstract] |
Monday, March 5, 2007 1:03PM - 1:15PM |
B27.00010: Including spin-orbit coupling in materials-specific studies of spin transport. A.A. Starikov, P.J. Kelly Spin-orbit coupling (SOC) plays a crucial role in magnetoelectronics: it is the origin of anisotropic magneto-resistance (AMR), prevents half-metallic ferromagnets from having 100\% spin polarization, gives rise to spin-flip scattering which ultimately destroys the spin polarization of a current in non-magnetic materials - to mention but a few of its effects. Nevertheless, it has been virtually ignored in theoretical transport studies. To redress this neglect, we have developed a method based upon Linearized Muffin-Tin-Orbitals suitable for studying spin-dependent transport in nanostructures which includes SOC and provides a framework for modelling layered magnetic systems with non-collinear magnetizations. As a first application and test of the method, we study the AMR effect in ferromagnetic alloys. [Preview Abstract] |
Monday, March 5, 2007 1:15PM - 1:27PM |
B27.00011: Modeling the deformation of materials with stochastic fractal microstructure M.A. Soare, R.C. Picu Many materials with heterogeneous multiscale fractal structure are found in nature. Examples include biological tissues and bone, some rock such as sandstones, and aero-gels. In such materials the amount of geometrical detail observed in the microstructure increases from scale to scale in a self-similar manner, they lack characteristic length scales and the Hausdorff dimension is smaller than that of the embedding space. Furthermore, the microstructure is multiscale and stochastic, in the sense that the generating operators that map the geometry from one scale to the next are stochastic. In this work, we develop a method by which boundary value problems can be solved for these complex multiscale materials with minimal computational effort. Use is made of the scaling properties of the geometry and of stochastic finite elements in which the solution is approximated using chaos polynomials. The talk will review the formulation and a number of examples used for verification. [Preview Abstract] |
Monday, March 5, 2007 1:27PM - 1:39PM |
B27.00012: Binding energies of CO$_{2}$ with some ionic liquids William Eucker, John Bendler Room temperature ionic liquids (RTILs), a novel class of materials with negligible vapor pressures and potentiality as benign solvents, may be an ideal chemical for carbon dioxide (CO$_{2})$ gas sequestration. \textit{Ab initio} computational modeling was used to investigate the molecular interactions of simple RTIL anions hexafluorophosphate (PF$_{6}^{-})$ and tetrafluoroborate (BF$_{4}^{-})$ with CO$_{2}$. Electronic potential energy surface (PES) scans of a comprehensive sampling of 1:1 anion-CO$_{2}$ orientations were computed using Spartan '02 with Dunning's correlation consistent basis sets. Qualitatively, the PES scans yielded deeper, more numerous and radially closer active sites surrounding BF$_{4}^{-}$ anion as compared with the PF$_{6}^{-}$ anion. Quantitatively, the binding energies of 17.87 kJ/mol and 25.24 kJ/mol were extracted from the identified global energy minima for the PF$_{6}^{-}$ and BF$_{4}^{-}$ systems, respectively. The smaller BF$_{4}^{-}$ anion was concluded to bind more strongly to the CO$_{2}$. However, literature-reported experimental Henry's law constants for CO$_{2}$ dissolved in imidizolium based RTILs show greater gas solvation in the PF$_{6}^{-}$ system. The discrepancy between the energetics calculation results and the experimental solvation data will be discussed. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700