Bulletin of the American Physical Society
APS March Meeting 2018
Volume 63, Number 1
Monday–Friday, March 5–9, 2018; Los Angeles, California
Session P12: Computational Materials Design - Databases and ToolsFocus
|
Hide Abstracts |
Sponsoring Units: DMP DCOMP Chair: Shyue Ping Ong, University of California, San Diego Room: LACC 303B |
Wednesday, March 7, 2018 2:30PM - 3:06PM |
P12.00001: Materials Informatics for the Discovery of Novel 2D Materials Invited Speaker: Richard Hennig The rapid rise of novel single-layer materials presents the exciting opportunity for materials science to explore an entirely new class of materials. This comes at the time when mature computational methods provide the predictive capability to enable the computational discovery, characterization, and design of single-layer materials and provide the needed input and guidance to experimental studies. I will present our data-mining, chemical substitution, and evolutionary algorithm approaches to identify novel 2D materials with low formation energies and show how unexpected structures emerge when a material is reduced to sub-nanometers in thickness. To identify 2D materials that can be synthesized by exfoliation of bulk materials, we searched the Materials Project crystal structure database for materials possessing layered motifs in their crystal structures using a topology-scaling algorithm. The algorithm identifies and measures the sizes of bonded atomic clusters in a structure’s unit cell, and determines their scaling with cell size. The search yielded 680 monolayers with exfoliation energies below those of already-existent 2D materials. These materials guide future experimental synthesis efforts. Among the 2D materials, we find that for several 2D transition-metal chalcogenide compounds ferromagnetic order emerges at temperatures accessible to experiments. Calculations of the magnetic anisotropy show that many of the magnetic 2D materials exhibit an easy-plane for the magnetic moment and hence a Berezinsky-Kosterlitz-Thouless transition to a magnetically ordered low-temperature phase. A few 2D materials display an easy magnetization axis and thus an actual ferromagnetic ground state. Furthermore, we identify a family of three magnetic 2D materials with half-metallic band structures. Their purely spin-polarized currents and dispersive interlayer interactions should make these materials useful for 2D spin valves and other spintronic applications. These new 2D materials provide the opportunity to investigate the interplay of magnetic order and reduced dimensionality and may provide materials suitable for optoelectronic and spintronic applications. The structures and other calculated data for all 2D materials are available in the MaterialsWeb database at https://materialsweb.org. |
Wednesday, March 7, 2018 3:06PM - 3:18PM |
P12.00002: PERTURBO: An open-source software for the accelerated discovery of electron scattering and dynamical processes in materials Luis Agapito, I-Te Lu, Marco Bernardi Solid-state technologies critically depend on charge carrier dynamics. Electrons are scattered by other quasiparticles and excitations present in the material (e.g., phonons, defects, and photons). Computing the rate of these scattering processes and their impact on carrier dynamics requires novel parallel algorithms with extensive memory requirements. |
Wednesday, March 7, 2018 3:18PM - 3:30PM |
P12.00003: Quality Control of Numerical Settings for DFT Calculations and Materials Databases Christian Carbogno, Kristian Thygesen, Björn Bieniek, Claudia Draxl, Luca Ghiringhelli, Andris Gulans, Oliver Hofmann, Karsten Jacobsen, Sven Lubeck, Jens Mortensen, Mikkel Strange, Elisabeth Wruss, Matthias Scheffler Density-functional theory (DFT) has become an invaluable tool in materials science. Whereas the precision of different approaches has been scrutinized for the PBE functional using extremely accurate numerical settings [1], little is yet known about code- and method-specific errors that arise under more commonly used settings. Recently, this has become a severe issue, since it prevents repurposing publicly available DFT data created using different settings and/or codes. To overcome this, we study the convergence of different properties (geometries, total and relative energies) in four different DFT codes (exciting, FHI-aims, GPAW, VASP) for typical settings. Specifically, we discuss relative and absolute errors as a function of the numerical settings, e.g., basis sets and k-grids, for 71 elemental solids [1]. Using this data, we propose analytical models that allow for reliable error estimates for any compound, as we explicitly demonstrate for binary and ternary solids. We discuss the extensibility of our approach towards more complex materials properties and its applicability in computational materials databases. |
Wednesday, March 7, 2018 3:30PM - 3:42PM |
P12.00004: PAOFLOW: A utility to construct and operate on ab initio Hamiltonians from the Projections of electronic wavefunctions on Atomic Orbital bases (PAO), including characterization of topological materials Frank Cerasoli, Marco Buongiorno Nardelli, Marcio Costa, Stefano Curtarolo, Riccardo De Gennaro, Marco Fornari, Laalitha Liyanage, Andrew Supka, Haihang Wang Rapid progression of computational technology during the past half century allows material sceintists to predict electronic and bulk properties of increasingly complex systems. PAOFLOW is a utility for the analysis and characterization of these properties from the output of any electronic structure calculation in Density Functional Theory (DFT). By exploiting an efficient procedure to project the full plane-wave solution of the original DFT Hamiltonian to a reduced space of atomic orbitals (Projection on Atomic Orbitals, or PAO) we can calculate a plethora of quantities with a negligible computational cost. This code, written entirely in Python under GPL 3.0 or later, opens the way to the high-throughput computational characterization of materials at an unprecedented scale. |
Wednesday, March 7, 2018 3:42PM - 3:54PM |
P12.00005: GBMaker: an efficient and open-source Python library for making grain boundaries Jianli Cheng, Jian Luo, Kesong Yang In this talk, we will present an efficient and open-source Python library for generating periodic boundary structures, Grain Boundary Maker (GBMaker). This software is designed to construct various grain boundary structures from a general cubic and non-cubic bulk crystal structure. A convenient command line tool has also been provided to enable easy and fast construction of tilt and twist boundaries by providing parameters including the degree of fit (Σ), rotation axis, grain boundary plane and initial bulk crystal structure. Our software can greatly help accelerate the computational and theoretical investigation of grain boundary structures. |
Wednesday, March 7, 2018 3:54PM - 4:06PM |
P12.00006: Accelerated modeling of electron transport using Bloch waves Maarten Van de Put, Massimo Fischetti, William Vandenberghe In computational condensed matter, the modeling of quantum transport in nanoscaled structures has received considerable attention, driving the development of ever smaller and faster electronics. |
Wednesday, March 7, 2018 4:06PM - 4:18PM |
P12.00007: Novel simulation methodologies for accelerated knowledge-based materials design. Igor Abrikosov Ab initio electronic structure theory is known as a useful tool for prediction of materials properties. However, until recently majority of simulations dealt with calculations at zero temperature that employ local or semi-local functionals. We present new methodological solutions, which go beyond this approach and explicitly take into account many-electron and finite temperature effects. Basic ideas behind the generalization of the Temperature Dependent Effective Potential (TDEP) method for the treatment of solid solutions [1], as well as its combination with Disordered Local Moment Molecular Dynamics (DLM-MD) [2] are introduced, and their capability is demonstrated in simulations of multicomponent nitride alloys for hard-coating applications. We show that state-of-the-art computer simulations can be used for accelerated knowledge-based materials design [3]. |
Wednesday, March 7, 2018 4:18PM - 4:30PM |
P12.00008: High-throughput Density-Functional Perturbation Theory phonons for inorganic materials Guido Petretto, Shyam Dwaraknath, Henrique Miranda, Michiel van Setten, Matteo Giantomassi, Donald Winston, Patrick Huck, Xavier Gonze, Kristin Persson, Geoffroy Hautier, Gian-Marco Rignanese The knowledge of the vibrational properties of a material is of key importance in understanding a variety of physical phenomena. For instance, the phonon spectrum of a material contains all the information to describe the dynamic of its constituent atoms in the harmonic approximation. However, detailed exper- imental values of the full phonon spectrum are available for a limited number of materials and this hinders the possibility of performing large scale analysis of vibrational properties and their derived quantities. Moving to a larger scale with density functional perturbation calculations, however, requires the presence of a robust framework to handle this challenging task. In light of this, we automatized the phonon calculation and applied the result to the analysis of the convergence trends for several materials. This allowed to identify and tackle some common problems emerging in this kind of simulations and to lay out the basis to obtain reliable phonon band structures from high-throughput calculations. We then ap- plied our framework in order to calculate the full phonon band structures and density of states for a large number of semiconductor compounds, as long as their derived quantities. |
Wednesday, March 7, 2018 4:30PM - 4:42PM |
P12.00009: Genarris: Random Generation of Molecular Crystal Structures and Fast Screening with a Harris Approximation Xiayue Li, Farren Curtis, Timothy Rose, Christoph Schober, Alvaro Vazquez-Mayagoitia, Harald Oberhofer, Noa Marom We present Genarris, a Python package that performs configuration space screening for molecular crystals of rigid molecules by random sampling with physical constraints. For fast total energy evaluations Genarris employs a Harris approximation, whereby the total density of a molecular crystal is constructed via superposition of single molecule densities. Dispersion-inclusive density functional theory (DFT) is then used to evaluate the total energy of the Harris density without performing a self-consistency cycle. Genarris uses machine learning for clustering, based on a relative coordinate descriptor (RCD) developed specifically for molecular crystals, which we find to be robust in identifying packing motif similarity. Genarris offers three workflows to screen a raw pool of random structures via different sequences of successive clustering and filtering steps: the “Rigorous” workflow is an exhaustive exploration of the potential energy landscape, the “Energy” workflow produces a set of low energy structures, and the “Diverse” workflow produces a maximally diverse set of structures. The usage of Genarris is demonstrated for three test cases of past blind test targets. |
Wednesday, March 7, 2018 4:42PM - 4:54PM |
P12.00010: Raising the bar for accessibility and sustainability of data published in scientific papers Marco Govoni, Milson Munakami, Aditya Tanikanti, Jonathan Skone, Hakizumwami Runesha, Juan De Pablo, Giulia Galli The availability of data presented in scientific papers is often hindered by the lack of direct links between published results and the datasets used to generate them. In the case of papers published by the physics, chemistry and materials science communities, such links may be a complex combination of connections to large and heterogeneous datasets. We developed a platform for the dissemination and reproducibility of data on a per-publication basis. We envision each scientific paper to be complemented by electronic notebooks that describe dataset manipulations, and with metadata available to describe provenance of all used codes and experiments. We will discuss the main pillars of the infrastructure developed to make data searchable and shareable, and metadata available. |
Wednesday, March 7, 2018 4:54PM - 5:06PM |
P12.00011: High-throughput hybrid-functional DFT investigation of materials band gaps and formation energies Mohan Liu, Vinay Hegde, Christopher Wolverton The improvement in calculated materials properties with Heyd-Scuseria-Ernzerhof (HSE) hybrid-functional method over conventional density functional theory (DFT) is often not systematic and has yet to be evaluated thoroughly for different classes of materials. As HSE calculations are significantly more expensive than conventional DFT calculations with the semi-local Perdew-Burke-Ernzerhof (PBE) exchange correlation, it is highly desirable to have the ability to determine a priori which method would provide more accurate results for a certain electronic or thermodynamic property. In this contribution, we present a high-throughput hybrid-functional (HSE) DFT database of band gaps and formation energies for over a thousand materials. Comparing with experimental data, we found that materials band gaps obtained through HSE are more accurate than those from PBE. In general, PBE tends to underestimate the band gaps, while HSE significantly decreases the average prediction error from ~1.1 eV down to ~0.5 eV. For formation energies, there is a linear correlation between formation energies calculate by HSE and PBE, where PBE is systematically lower in magnitude than HSE. |
Wednesday, March 7, 2018 5:06PM - 5:18PM |
P12.00012: An Electronic Transport Properties Database From High-Throughput Ab-initio Computations Francesco Ricci, Wei Chen, Umut Aydemir, Jeff Snyder, Gian-Marco Rignanese, Anubhav Jain, Geoffroy Hautier Nowadays the state-of-the-art DFT codes and the high-throughput (HT) frameworks allow us to compute materials properties at a large scale. The Material Project (MP) is one of the biggest project that aims to compute and share structures and properties of materials. As recently made for elastic and piezoelectric tensors, we will present a large and freely accessible data set of transport properties as effective mass and Seebeck coefficient. This transport data has been computed on top of energy band structures available in MP, using the well-known BoltzTraP code inserted in a HT framework. Given the importance of electronic transport properties, the whole community of material science researcher will benefit from this database. We will present the work flow to obtain the data and the data set. We will also study some of the correlation between transport properties and present applications in the field of transparent conducting oxides and thermoelectric materials. |
Wednesday, March 7, 2018 5:18PM - 5:30PM |
P12.00013: Automatic Generation of Nonpolar Stoichiometric Slab Models Requiring Surface Reconstruction Yoyo Hinuma The slab-and-vacuum model is an indispensable tool to derive energetic, atomistic, and electronic properties of surfaces of materials with first principles codes that employ three-dimensional periodic boundary conditions. Polar and nonpolar slabs require different levels of treatment, as any polar instability must be compensated on a case-by-case basis in the former. Some slabs can be made nonpolar and stoichiometric by simply cleaving from bulk (nonpolar type A and B), while in other cases nonpolar and stoichoimetric slabs can only be attained with surface reconstruction after cleavage (nonpolar type C) [Hinuma et al., Comp. Mater. Sci. 113, 221 (2016)]. In this talk, I will show a procedure exploiting on the crystallographic concept of isometry to automatically reconstruct the topmost and bottommost layers of a nonpolar type C slab to attain stoichiometry. Although this method is not applicable to all nonpolar type C surfaces, automatic generation of nonpolar slabs of many crystals, including perovskite and spinel structures, has now become possible, paving the way toward further understanding of surface physics. |
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