Bulletin of the American Physical Society
APS March Meeting 2010
Volume 55, Number 2
Monday–Friday, March 15–19, 2010; Portland, Oregon
Session P28: Focus Session: New Frontiers in Electronic Structure Theory III |
Hide Abstracts |
Sponsoring Units: DCP Chair: Martin Head-Gordon, University of California, Berkeley Room: C124 |
Wednesday, March 17, 2010 8:00AM - 8:36AM |
P28.00001: Advances in Orbital-Free Density Functional Theory: Physics and Algorithms Invited Speaker: Orbital-free density functional theory (OFDFT) is a first principles quantum mechanics method that can scale linearly with system size by solving directly for the electron density instead of introducing an auxiliary set of one-electron orbitals as is done in conventional Kohn-Sham (KS) DFT. Orbitals must be kept orthonormal; imposing this constraint involves a cubically scaling step. KSDFT can be made to scale linearly beyond a crossover point within a localized orbital framework and hence molecules and insulators can be made to scale linearly within KSDFT. Metals generally do not exhibit linear scaling within KSDFT due to their inherently delocalized electronic structure. OFDFT offers an alternative scheme by introducing a kinetic energy density functional (KEDF) and local electron-ion pseudopotentials (LPSs). Thus the accuracy of OFDFT depends on the representation of these two terms. We now have a routine tool for constructing accurate LPSs by inverting the KS equations for bulk crystals. These BLPSs are validated against accurate nonlocal PSs within KSDFT. A decade ago, we reported a nonlocal KEDF that accurately captures the physics of nearly-free-electron-like metals. Here we report a new nonlocal KEDF that explicitly contains the physics required to describe semiconductors. Extensive tests on silicon and a variety of compound semiconductors reveal this new KEDF to be accurate for many properties, thus opening up the door to accurate OFDFT calculations on semiconductors. We also report the systematic elimination of bottlenecks within our OFDFT code that render the entire algorithm linear scaling for all system sizes (no crossover point). With parallelization then introduced via domain decomposition, quantum mechanical simulations of metal samples containing up to 1 million atoms have been demonstrated using a modest number of processors. With this new capability we are studying mesoscale features that control mechanical properties of Al and Mg alloys. [Preview Abstract] |
Wednesday, March 17, 2010 8:36AM - 8:48AM |
P28.00002: Charge-transfer excited states using a constrained density functional method Marco Olguin, Rajendra Zope, Tunna Baruah Growing interest in characterizing charge-transfer (CT) processes inherent in many chemical interactions, such as metal surface-adsorbate catalysis or light harvesting processes, has prompted a demand for practical electronic structure methods with sufficient accuracy to describe the excited-states of large systems. The widely used time-dependent density functional formalism shows limitations in describing CT excited-states. In this study, we employ a recently developed constrained excited-state method to estimate the energies of singly excited particle-hole states. The applied constraint in this approach is the orthogonality between the ground and excited-states. The expense involved in the calculation of each excited state is comparable or slightly less than that for the ground state. A comparison of the gas-phase CT excitation energies for a set of aromatic donor with TCNE acceptor show excellent agreement with experimental values. Its application to the CT excited states for a light harvesting C60-porphyrin-beta-carotene triad demonstrates the applicability of this approach to large systems. The performance appraisal of this method for a large number of excited states including core excitation will be presented and further improvements will be discussed. [Preview Abstract] |
Wednesday, March 17, 2010 8:48AM - 9:00AM |
P28.00003: Phase-Space Density Functional Theory: A Quasiclassical Approach Izabela Raczkowska, Arun Rajam, Neepa Maitra In time-dependent density functional theory, one obtains all observables pertaining to an interacting electronic system from one that is non-interacting, in principle exactly. In practice approximations are needed for the exchange-correlation potential and also observable-functionals. Although presently used functionals have been successful for many applications, they perform poorly for several phenomena of interest eg. momentum distributions, electronic quantum control problems, and where there is strong memory-dependence. We propose and explore a possible solution to these problems by extending TDDFT to phase-space densities, and discuss dynamical quasiclassical and semiclassical approximations for the correlation functional in phase-space. [Preview Abstract] |
Wednesday, March 17, 2010 9:00AM - 9:12AM |
P28.00004: The unreasonable accuracy of asymptotic expansions: Why density functional theory works Kieron Burke, Attila Cangi, Donghyung Lee, Peter Elliott I will discuss why local and semilocal approximations work as well as they do (and why they fail where they fail). [Preview Abstract] |
Wednesday, March 17, 2010 9:12AM - 9:24AM |
P28.00005: Non-Koopmans correction to the convexity and inaccuracy of orbital levels in local and semilocal density-functional theories Ismaila Dabo, Andrea Ferretti, Nicolas Poilvert, Nicola Marzari, Matteo Cococcioni Local and semilocal density-functional approximations provide excellent predictions for systems with non-fractional electron occupations. However, such theories considerably overestimate the stability of systems with fractionally occupied orbitals. This fundamental deficiency, which arises from the convexity of approximate functionals and the presence of self-interaction, is responsible for a number of qualitative and quantitative failures, which pervade all aspects of electronic-structure predictions, ranging from electron transfer, electron transport, electronic polarization to molecular dissociation and adsorption. We focus here on the most immediate manifestation of the convexity problem, i.e., the inaccuracy of orbital energy levels. We demonstrate that errors in predicting orbital energies can be eliminated by introducing a non-Koopmans (NK) correction based upon the satisfaction of Koopmans' theorem, which identifies orbital energies with opposite removal energies in the frozen orbital approximation. We demonstrate the remarkable performance of the NK approach in predicting orbital levels for a complete range of atoms and molecules. We then examine the accuracy of the NK correction for large polyatomic systems in which the effect of orbital relaxation is expected to be significant. [Preview Abstract] |
Wednesday, March 17, 2010 9:24AM - 10:00AM |
P28.00006: New models for mixing wavefunctions with density functional theory Invited Speaker: The recent realization that the ground-state correlation energy of the random phase approximation (RPA) is intimately connected to an approximate coupled cluster doubles (CCD) model [1], opens interesting avenues for mixing RPA with DFT [2]. I will also present a new constrained-pairing mean-field theory (CPMFT) that describes strong correlations quite accurately [3,4]. Dynamical correlation functionals for this model are feasible but they require the use of ``alternative'' densities. \\[4pt] [1] G. E. Scuseria, T. M. Henderson, and D. C. Sorensen, J. Chem. Phys. \textbf{129}, 231101 (2008). \\[0pt] [2] B. G. Janesko, T. M. Henderson, and G. E. Scuseria, J. Chem. Phys. \textbf{130}, 081105 (2009). \\[0pt] [3] T. Tsuchimochi and G. E. Scuseria, J. Chem. Phys. \textbf{131}, 121102 (2009). \\[0pt] [4] G. E. Scuseria and T. Tsuchimochi, J. Chem. Phys. \textbf{131}, 164119 (2009). [Preview Abstract] |
Wednesday, March 17, 2010 10:00AM - 10:12AM |
P28.00007: Implicit solvent model for density functional theory calculations Hatem Helal, Mike Payne, Arash A. Mostofi There has been a considerable amount of work in developing simulation techniques for processes in solution. Implicit solvation is one such method which successfully addresses the difficulty of achieving a reliable statistical average over the many solvent degrees of freedom, and this property makes it the prime candidate for use within efficient density functional theory (DFT) calculations on biological systems. The implicit solvation method replaces the complex arrangement of solvent molecules with a continuous polarizable medium with the intention of replicating the electrostatic response of the bulk solvent. The method we have implemented uses the electron density to describe the solvation cavity and thus departs from standard implicit solvation methods which are burdened by a large number of parameters to define the solvent cavity. We present the validation of this model through the calculation of energies of solvation, as well as an investigation of the solvation effect on NMR chemical shifts of small biologically relevant molecules. [Preview Abstract] |
Wednesday, March 17, 2010 10:12AM - 10:24AM |
P28.00008: Transition densities in time-dependent density functional theory Yonghui Li, Carsten Ullrich Real-space density-matrix analysis is a useful computational tool to visualize and interpret the induced charges and electron-hole coherences of electronic excitations in molecules. We extend this technique into the nonlinear, real-time domain and define the time-dependent transition densities in the context of time-dependent density-functional theory. This provides a real-time visualization tool for optical excitation processes in molecules, which will be illustrated for simple one-dimensional lattice model systems. Comparisons with numerically exact many-body benchmark solutions will be carried out by constructing the corresponding exact time-dependent Kohn-Sham transition density matrix. This work is supported by NSF Grant DMR-0553485 [Preview Abstract] |
Wednesday, March 17, 2010 10:24AM - 10:36AM |
P28.00009: Multi-domain decomposition method for real-time propagation of wave function Vladimir Goncharov, Kalman Varga Extension of electronic structure methods to ever-large systems is an important problem in computational material and bio sciences. We have developed a time-dependent density functional approach that is capable to describe electron dynamics in large molecular complexes and realistic nanostructures. For wave function propagation, the most expensive numeric operation is the evaluation of product of Hamiltonian matrix and wave function. We use a multi-domain decomposition to increase numerical efficiency of this operation that results in efficient real time propagation of wave function in real space. We currently perform extensive testing of the new system and plan to use it in calculations of optical absorption spectra of supra-molecular assemblies. Results will be presented for C60-Fe-porphyrin complex. [Preview Abstract] |
Wednesday, March 17, 2010 10:36AM - 10:48AM |
P28.00010: Orbital magnetoelectric response of insulating crystals Andrei Malashevich, Ivo Souza, Sinisa Coh, David Vanderbilt We calculate the orbital magnetoelectric polarizability (OMP) of a periodic insulator as the linear orbital magnetization response to a homogeneous electric field. We begin by considering the orbital magnetization (OM) in a finite field, and find that it can be written as a sum of three terms, one of which has no counterpart at zero field. The extra contribution is parallel to the electric field and is a multivalued quantity, only defined up to a field-dependent quantum. The full OM expression can be implemented in {\it ab-initio} codes, allowing to calculate the OMP by finite differences. Alternatively, linear-response techniques may be used, and for that purpose we obtain an expression directly for the OMP, by taking analytically the field derivative of the OM formula. The resulting expression has two types of terms: one which is expressed solely in terms of the unperturbed valence wavefunctions, and another coming from the first-order change in the wavefunctions. In normal insulators both are present, while in strong topological insulators only the former type is present. The full expression for the OMP tensor is verified numerically for a 3D tight-binding model of a normal insulator with broken inversion and time-reversal symmetries, by comparing with the finite-field orbital magnetization calculated for both finite and periodic samples. [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