Bulletin of the American Physical Society
APS March Meeting 2018
Monday–Friday, March 5–9, 2018; Los Angeles, California
Session L18: Electronic Structure Methods |
Hide Abstracts |
Sponsoring Units: DCOMP DMP Chair: Efthimios Kaxiras, Harvard Univ Room: LACC 306B |
Wednesday, March 7, 2018 11:15AM - 11:27AM |
L18.00001: Geometrical Observables of the Electronic Ground State Raffaele Resta , Antimo Marrazzo In a crystalline material the geometrical observables obtain either as Brillouin-zone integrals (in insulators), or as Fermi-volume integrals (in metals). In two cases the integrand is gauge-dependent and the bulk observables are defined modulo a "quantum": these are polarization and the "axion" term in magnetoelectric response. For bounded samples, the actual values of these two observables depend on the details of the sample termination. Remarkably, four other cases belong to a very different class, in that the integrand is gauge invariant, and the observable is independent of the sample boundary. They are: Drude weight (in metals), orbital magnetization, gauge-invariant quadratic spread (in insulators), and anomalous Hall conductivity. In our work we first provide a common k-space formalism for the four observables; then we also show that all of them admit a dual formulation in r space, and are local in character. Our local formulation allows to address the above four observables in noncrystalline samples, as well as in bounded and/or inhomogeneous samples. Simulations on paradigmatic cases validate our approach. |
Wednesday, March 7, 2018 11:27AM - 11:39AM |
L18.00002: Compressed Modes for Topological Insulators using Eigenspace Projection Bradley Magnetta , Vidvuds Ozolins , Jiatong Chen In general, Wannier functions are localized real space functions that are defined through a unitary transformation of eigenfunctions. Here we focus on a particular variant of Wannier functions, known as compressed modes (CMs), which are traditionally obtained from the minimization of the total energy plus an L1 regularization term. We demonstrate the difficulties that arise when calculating CMs for tight-binding models of topological insulators and suggest a new approach for calculating CMs by defining an objective functional that uses eigenspace projection. |
Wednesday, March 7, 2018 11:39AM - 11:51AM |
L18.00003: Benchmarking the performance of the SCAN meta-GGA functional for solid-state thermodynamics Eric Isaacs , Christopher Wolverton Constructed to satisfy all known exact constraints and appropriate norms for a semi-local density functional, the SCAN meta-GGA has shown early promise in the accurate description of electronic structure of molecules and solids [1]. One open question is how SCAN performs in describing the thermodynamics of solid-state materials. To answer this question, we perform an extensive benchmark of SCAN by computing the formation energies for a diverse group of nearly one thousand crystalline solids for which experimental values are available. While SCAN substantially decreases the formation energy errors for strongly bound solids (by around 50%) compared to the GGA level of theory, for intermetallic compounds we see no improvement. We find that SCAN leads to significantly more accurate structural parameters, mildly improved band gaps, and enhanced magnetism compared to GGA. Additionally, we present elemental chemical potential corrections for SCAN that are fit to achieve best agreement with experimental formation energies. |
Wednesday, March 7, 2018 11:51AM - 12:03PM |
L18.00004: Accurate Critical Pressures for Structural Phase Transitions of Group IV, III-V, and
II-VI Compounds from the SCAN Density Functional Chandra Shahi , Jianwei Sun , John Perdew The structural phase transition of a solid under pressure is a sensitive test of an approximate density functional. We have investigated the performance of the new meta generalized gradient approximation (meta-GGA) SCAN for the structural transitions of group IV, III-V, and III-VI compounds. SCAN performs better than the local density approximation (LDA) and generalized gradient approximation (GGA) PBE, yielding equilibrium pressures in reasonable agreement with experiment. For materials where calculations with computationally expensive methods like the hybrid functional HSE06 and the random phase approximation RPA have been reported, SCAN shows a comparable accuracy, at an acceptable computational cost. |
Wednesday, March 7, 2018 12:03PM - 12:15PM |
L18.00005: A SCAN perspective on the puzzle of the α ↔γ phase transition of Ce Abhirup Patra , Chandra Shahi , Puskar Bhattarai , John Perdew The recently developed SCAN meta-GGA and its van der Waals (vdW)- corrected form SCAN+rVV10 have shown remarkably good performance for different physical and chemical properties [1,2,3]. The interplay between self-interaction and vdW corrections can be crucial for transition metal oxides [4]. We recently studied other challenging problems, including the isostructural phase transition (α↔γ) of Ce. Both the α and γ phases have fcc structure, but with significant volume difference due to spatially localized 4f electrons in the γ phase. LDA and PBE capture only the alpha phase, while SCAN shows an energy minimum for α and a shoulder for γ phase. The magnetic moment for γ -Ce using SCAN is in good agreement with that of the PBE0 hybrid functional. |
Wednesday, March 7, 2018 12:15PM - 12:27PM |
L18.00006: Importance of van der Waals interaction on structural, vibrational, and thermodynamics properties of NaCl Lucy Assali , Michel Marcondes , Renata Wentzcovitch Thermal equations of state (EoS) are essential in several scientific domains. However, experimental determination of EoS parameters may be limited at extreme conditions, therefore, ab initio calculations have become an important method to obtain them. Density Functional Theory (DFT) and its extensions with various degrees of approximations for the exchange and correlation (XC) energy is the method of choice, but large errors in the EoS parameters are still common. The alkali halides have been problematic from the onset of this field, and the quest for appropriate DFT functionals for such ionic and relatively weakly bonded systems has remained an active topic of research. Here we use DFT + van der Waals functionals to calculate vibrational properties, thermal EoS, thermodynamic properties, and the B1 to B2 phase boundary of NaCl. Our results reveal i) a remarkable improvement over the performance of standard Local Density Approximation and Generalized Gradient Approximation functionals for all these properties and phase transition boundary, as well as ii) great sensitivity of anharmonic effects on the choice of XC functional. |
Wednesday, March 7, 2018 12:27PM - 12:39PM |
L18.00007: Correlation Energy of the Uniform Electron Gas Revisited Puskar Bhattarai , Chandra Shahi , Abhirup Patra , John Perdew Analytic parametrizations of the correlation energy per particle of a uniform electron gas are the common ingredient of nearly all density functional approximations. Standard parametrizations such as Vosko-Wilk-Nusair (VWN81), Perdew-Zunger (PZ81), and Perdew-Wang (PW92) are fitted to the Quantum Monte Carlo (QMC) results of Ceperley and Alder (1980), with additional information from the exact high- and low-density limits. They agree well with one another in the spin-unpolarized and fully polarized limits, but for density parameter r_{s}=0.5 they are more negative by about 0.1 eV (5 to 10%) than the QMC results of Spink, Needs, and Drummond (2013). Another approach, based almost entirely on satisfaction of all known exact constraints on the high- and low-density limits, was proposed by Sun et al. PRB 81, 085123 (2010) in their Density Parameter Interpolation (DPI). Here we correct a high-density-expansion coefficient for full spin polarization used in the original DPI. Our new coefficient agrees with that of Ref. [1]. We find that DPI still agrees more closely with VWN, PZ, and PW than with the QMC of Spink et al. The dependence upon relative spin polarization has also been studied. |
Wednesday, March 7, 2018 12:39PM - 12:51PM |
L18.00008: Correlation matrix renormalization theory for correlated-electron materials and application to crystalline phases of atomic hydrogen Cai-Zhuang Wang , Xin Zhao , Jun Liu , Yongxin Yao , Kai-Ming Ho Developing accurate and computationally efficient methods to calculate the electronic structure and total energy of correlated-electron materials has been a challenging task in condensed matter physics and materials science. We have developed a correlation matrix renormalization (CMR) method which does not assume any empirical Coulomb interaction U parameters and does not have double counting problems in the total energy calculation. It is demonstrated to be accurate in describing the electron correlations in both the molecules and periodic solid systems. Using linear hydrogen chain as benchmark, we show that the results from the CMR method compare very well with those obtained recently by accurate quantum Monte Carlo (QMC) calculations. We also study the equation of states of three-dimensional crystalline phases of atomic hydrogen. The results from the CMR method agree much better with the available QMC data than those from density functional theory (DFT) and Hartree-Fock (HF) calculations. |
Wednesday, March 7, 2018 12:51PM - 1:03PM |
L18.00009: A full-potential approach to the solution of core states Ziyin Liu , Xianglin Liu , Yang Wang For an ab intio, all-electron, electronic structure calculation method, e.g. KKR, LMTO, calculating the core states associated with each atom in unit cell is a necessary procedure during its self-consistent iterations. In this procedure, the energy eigenvalue and the electron density of the core states are obtained by solving the Kohn-Sham equation, with boundary conditions that the wavefunction solutions become zero beyond a certain distance from the nucleus, e.g., the muffin-tin radius. Despite that a full-potential method is required for solving the valence states when muffin-tin potential approximation becomes invalid, the core states are still solved with the potential in muffin-tin form. In this presentation, we show a full-potential method, based on scattering theory, for calculating the core states. In this approach, we developed a fast method for searching the poles of S-matrix to find the core state energy eigenvalues, and we applied a Green function technique to calculate the electron density associated with the core states. As a result, the solutions for both valence and core states correspond to the same one-electron potential. This method enables proper treatment of shallow core states, which appear in materials under high pressure with high lattice distortions. |
Wednesday, March 7, 2018 1:03PM - 1:15PM |
L18.00010: All-electron explicitly correlated calculations of S- and P-states of beryllium atom Sergiy Bubin We report new benchmark calculations of several lowest S- and P-states of Be atom carried out in the framework of the variational method that employs large and highly optimized all-particle explicitly correlated Gaussian basis sets. In the calculations we include the effects due to finite nuclear mass as well as the leading relativistic and QED corrections. A comparison of the computed transition energies with experimentally derived values is made and the sources of the remaining discrepancies are discussed. |
Wednesday, March 7, 2018 1:15PM - 1:27PM |
L18.00011: Understanding excited-states of light-harvesting chromophores with ab initio many-body perturbation theory Samia Hamed , Milan Delor , Fabien Bruneval , Naomi Ginsberg , Jeffrey Neaton Biomimetic light-harvesting systems provide a rich platform for distilling the photophysics underlying natural photosynthetic energy transfer processes. Deducing the mechanisms underpinning these dynamics requires an accurate characterization of the excited states of a chromophore molecule coupled to a dynamic and heterogeneous environment that can lead to complex non-radiative decay pathways. Here, we use time-dependent density functional theory (TDDFT) and ab initio many-body perturbation theory with the GW plus Bethe-Salpeter equation (GW-BSE) approach to predict the excited states and intermolecular coupling of chromophores. We have previously shown GW-BSE is capable of improving on TDDFT for simpler systems that feature differential electron correlation between ground and excited states. Additionally, we identify factors that give rise to the time scales observed in transient absorption experiments and those that yield the desired photodynamics and resulting energy transfer properties. |
Wednesday, March 7, 2018 1:27PM - 1:39PM |
L18.00012: First-principles Studies of Trapping States Associated with Impurities and Native Defects in Halide Scintillator Phosphor Materials and their Impact on Optical and Gamma Ray Detection Properties Andrew Canning , Mauro Del Ben , E. Bourret-Courchesne , Gregory Bizarri Halide scintillator phosphors doped with Ce or Eu are amongst the brightest known new gamma ray detector materials (e.g. LaBr_{3}:Ce, BaBrI:Eu, SrI_{2}:Eu). The complete scintillation process in these materials is poorly understood and in particular the role of trapping states associated with impurities and native defects. We have performed first principles studies based on GGA, hybrid functionals and the GW method in tandem with experiments to understand the role of defect states in improving or degrading the gamma ray detection properties of this class of materials. In particular we have looked at hole, electron and exciton trapping at defect states and their mobility in terms of transfer to the Ce or Eu activator and how that can explain the very different optical and gamma ray detection properties of this class of scintillator materials. |
Wednesday, March 7, 2018 1:39PM - 1:51PM |
L18.00013: GW Computations via GPU-GPU Message Passing: A path to High-throughput Nicholas Thompson , Jie Zhang The GW approximation uses a screened Coulomb interaction to calculate the self-energy of a system of interacting electrons. The resulting band structures and absorption spectra compare favorably with experiment, but the computations are currently too expensive to be used in high-throughput ab initio electronic structure calculations. In this talk we describe our implementation of the GW approximation, which uses GPU-GPU interprocess communication for acceleration, and discuss its advantages over traditional MPI code. In addition, we describe our "ATLAS"-style method of optimizing the code to prevent warp divergence on arbitrary CUDA-enabled GPU architectures, and end with an estimate of when this method will be cost-effective for high-throughput calculations assuming continued exponential growth in CUDA cores per GPU. |
Wednesday, March 7, 2018 1:51PM - 2:03PM |
L18.00014: QSGW+G method and its application to FeSb_{2} Yongxin Yao , Sangkook Choi , Nicola Lanata , Walber Brito , Gabriel Kotliar We present the quasi-particle self-consistent GW plus Gutzwiller rotationally invariant slave-boson (QSGW+G) method as a simplified version of QSGW + dynamical mean-field theory (DMFT). The application of QSGW+G to the thermoelectric material FeSb_{2} yields an electronic structure consistent with QSGW+DMFT, albeit with some difference in fine details. It further reveils that the indirect band gap size is much more sensitive to Hund's coupling J rather than Hubbard interaction U. Most importantly, QSGW+G generates the results in a tiny fraction of time needed by QSGW+DMFT, rendering it an efficient screening method before higher-level QSGW+DMFT calculations in search of real correlated functional materials. |
Wednesday, March 7, 2018 2:03PM - 2:15PM |
L18.00015: Roles of Hydrogen Partial Pressure in Controlled Sulfidation and Nucleation Process of Molybdenum Oxide* Chunyang Sheng , Sungwook Hong , Aiichiro Nakano , Rajiv Kalia , Priya Vashishta Molybdenum disulfide (MoS_{2}), a direct-bandgap 2 dimensional material, is a promising candidate for future electronics applications due to its unique electronic properties, for which thorough understanding of synthesis processes is indispensable. Experimental studies on the growth of MoS_{2} via chemical vapor deposition (CVD) revealed that single-layer MoS_{2} nanocrystals growth could be controlled by gas composition in reaction chamber. Here, we use quantum and reactive molecular dynamics simulations to investigate the effects of varying H_{2} partial pressure in the CVD growth process of MoS_{2}. Simulation results reveal key hydrogen-catalyzed reaction pathways and intermediate products. We also quantify the effects of H_{2} composition on the sulfidation and nucleation rates. These atomistic mechanisms not only explain experimental results but also shed light on controlled growth of MoS_{2} monolayers. |
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. |
© 2018 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
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700