Bulletin of the American Physical Society
APS March Meeting 2017
Volume 62, Number 4
Monday–Friday, March 13–17, 2017; New Orleans, Louisiana
Session K33: Computational Discovery and Design of Novel Materials VIIIFocus
|
Hide Abstracts |
Sponsoring Units: DMP DCOMP Chair: Anatole von Lilienfeld, University of Basel Room: 296 |
Wednesday, March 15, 2017 8:00AM - 8:36AM |
K33.00001: Computational Discovery of New Materials Under Pressure Invited Speaker: Eva Zurek The pressure variable opens the door towards the synthesis of materials with unique properties, ie. superconductivity, hydrogen storage media, high-energy density and superhard materials, to name a few. Indeed, recently superconductivity has been observed below 203 K and 103 K in samples of compressed sulfur dihydride and phosphine, respectively. Under pressure elements that would not normally combine may form stable compounds, or may mix in novel proportions. As a result using our chemical intuition developed at 1 atm to theoretically predict stable phases is bound to fail. In order to enable our search for superconducting hydrogen-rich systems under pressure, we have developed XtalOpt, an open-source evolutionary algorithm for crystal structure prediction. New advances in XtalOpt that enable the prediction of unit cells with greater complexity will be described. XtalOpt has been employed to find the most stable structures of hydrides with unique stoichiometries under pressure. The electronic structure and bonding of the predicted phases has been analyzed by detailed first-principles calculations based on density functional theory. The results of our computational experiments are helping us to build chemical and physical intuition for compressed solids. [Preview Abstract] |
Wednesday, March 15, 2017 8:36AM - 8:48AM |
K33.00002: The role of representation and training set selection for improved machine learning models of matter Bing Huang, O. Anatole von Lilienfeld Choice of representation and training set are fundamentally important in machine learning (ML) models of chemical and physical properties of matter. Based on the postulates of quantum mechanics we have developed a hierarchy of representations which meet uniqueness and target similarity criteria. To systematically control target similarity, we rely on interatomic many body expansions, as implemented in universal force-fields, including bags of sorted {\underline B}onding, {\underline A}ngular, and higher order terms (BA). Addition of higher order contributions systematically increases the predictive accuracy of the resulting BAML models. BAML predicts properties of out-of-sample molecules with unprecedented accuracy and speed.\footnote{Huang and von Lilienfeld, {\em J. Chem. Phys.} Comm. {\bf 145}, 161102 (2016)} To select optimal training sets we have developed a rational approach which results in ML models with very rapid error decay.\footnote{Huang and von Lilienfeld, in preparation (2016)} In combination with BAML based atomic representations, these ML models reach chemical accuracy for atomization energies ($\sim$1 kcal/mol) after training on reference results for only hundreds of chemical compounds. Our findings suggest a dramatic reduction in need for data. [Preview Abstract] |
Wednesday, March 15, 2017 8:48AM - 9:00AM |
K33.00003: Inverse problems in complex material design: Applications to non-crystalline solids Parthapratim Biswas, David Drabold, Stephen Elliott The design of complex amorphous materials is one of the fundamental problems in disordered condensed-matter science. While impressive developments of ab-initio simulation methods during the past several decades have brought tremendous success in understanding materials property from micro- to mesoscopic length scales, a major drawback is that they fail to incorporate existing knowledge of the materials in simulation methodologies. Since an essential feature of materials design is the synergy between experiment and theory, a properly developed approach to design materials should be able to exploit all available knowledge of the materials from measured experimental data. In this talk, we will address the design of complex disordered materials as an inverse problem involving experimental data and available empirical information. We show that the problem can be posed as a multi-objective non-convex optimization program, which can be addressed using a number of recently-developed bio-inspired global optimization techniques. In particular, we will discuss how a population-based stochastic search procedure can be used to determine the structure of non-crystalline solids (e.g. {\it a}-SiH, {\it a}-SiO$_2$, amorphous graphene, and Fe and Ni clusters). [Preview Abstract] |
Wednesday, March 15, 2017 9:00AM - 9:12AM |
K33.00004: An Integrated Computational and Data Environment to Support Multiscale Modeling of Soft Materials for the Materials Genome Initiative Frederick Phelan Jr., Thomas Rosch, Cheol Jeong, Brian Moroz, Sharief Youssef In this presentation, we describe the development of a computational ``workbench'' whose goal is to provide an integrated computational and data environment to support multiscale modeling of soft materials for the Materials Genome Initiative (MGI). The design has three essential elements: a modular program structure that supports the addition of new functionality through Python scripting and run-time plugins; a hierarchical data structure which enables unified representation of materials at different levels of granularity; finally, integration of the NIST Materials Data Curation System (MDCS) into the environment to support ontology based materials descriptions. The feature of the workbench which we emphasize in this presentation is coarse-graining. Coarse-graining techniques are an essential requirement for the design of soft materials, and are an active area of research across the soft matter community. We illustrate how the approach allows the integration of multiple coarse-graining techniques in a common environment to greater enable development, evaluation and comparison of new algorithms. Moreover, the environment meets the goals of the MGI by enabling automated curation of both upstream and downstream data in materials reference libraries which can be pushed or shared by various means. [Preview Abstract] |
Wednesday, March 15, 2017 9:12AM - 9:24AM |
K33.00005: A materials genomic approach to design of nucleating agents Alexander Bourque, Rebecca Locker, Gregory Rutledge Heterogeneous nucleation is frequently the first step in any system undergoing phase change. Nucleating agents are foreign materials introduced to modify and regulate this step. Control of nucleation kinetics permits modification of morphology in a broad range of crystallizable materials including those used for separators in batteries, fuels cells, etc. However, the mechanism(s) of action in heterogeneous nucleation are poorly understood, due in part to the small spatio-temporal scales over which nucleation occurs. Molecular simulation has been used to study the kinetics of homogeneous and, to a lesser extent, heterogeneous nucleation. We show that by systematically varying the intermolecular force field parameters that describe the nucleating agent, one can rapidly screen entire classes of materials to characterize both effectiveness and mechanism. The method is applied to the crystallization of n-pentacontane, a model surrogate for polyethylene, on the family of tetrahedrally coordinated crystals isomorphic with diamond and the family of 2D, hexagonally coordinated materials isomorphic with graphene. The induction time for heterogeneous nucleation is shown to depend strongly on crystallographic registry between the nucleating agent and the critical nucleus of the new phase, indicative of an epitaxial mechanism. Importantly, the severity of this registry requirement weakens with decreasing rigidity of the substrate and increasing strength of attraction to the nucleating agent. Employing this method, high throughput computational screening of nucleating agents becomes possible. [Preview Abstract] |
Wednesday, March 15, 2017 9:24AM - 9:36AM |
K33.00006: An alternative structure of TiO2 with higher energy valence band edge Sinisa Coh, Peter Y. Yu, Yuta Aoki, Susumu Saito, Steven G. Louie, Marvin L. Cohen We propose an alternative structure of TiO2 anatase that has a higher energy oxygen p-like valence band maximum than the pristine TiO2 anatase and thus has a much better alignment with the water splitting levels. This alternative structure is unique when considering a large subspace of possible structural distortions of TiO2 anatase. We propose two ways to access this state experimentally and argue that one of them might have been realized in the recently discovered so-called black TiO2. This work was supported by NSF Grant No. DMR-1508412 and the theory of Materials Program at the Lawrence Berkeley National Lab funded by the Director, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division, U.S. Department of Energy under Contract No. DE-AC02-05CH11231. Computational resources have been provided by the DOE at Lawrence Berkeley National Laboratory's NERSC facility. We acknowledge support from the MEXT Japan Elements Strategy Initiative to Form Core Research Center, MEXT Japan KAKENHI Grant No.25107005, and JSPS Grant No.14J11856. [Preview Abstract] |
Wednesday, March 15, 2017 9:36AM - 9:48AM |
K33.00007: Tuning Electronic Structure of Si$_{\mathrm{24}}$ by Doping and Strain Jiajun Linghu, Lei Shen, Ming Yang, YuanPing Feng Si24, a new allotrope of silicon with cmcm space group, has been synthesized recently and received much attention due to its quasi-direct band gap around 1.3 eV. Aiming to its potential application in solar cell device, in this study, we investigated the doping and strain effects on the electronic properties of Si$_{\mathrm{24}}$ using density-functional theory calculations with the hybrid functional. It is found that, Si$_{\mathrm{24}}$ can be easily doped as both p- and n-type semiconductors by III or V group elements. Among various potential dopants, B and P atoms are the most promising elements for the p-type and n-type doping in the Si$_{\mathrm{24}}$, respectively, because of their relatively low formation and ionization energies. More importantly, the incorporation of these two dopants would not introduce impurity bands within the band gap of Si$_{\mathrm{24}}$, but only cause a slightly narrowing of its band gap. Furthermore, through applying a small asymmetric compressive strain, the indirect band gap of pristine and doped Si$_{\mathrm{24}}$ can be tuned into the direct gap. It reveals the great potential in constructing novel Si$_{\mathrm{24}}$ based p-n junction which is highly desired for future industrial application in photovoltaic devices. [Preview Abstract] |
Wednesday, March 15, 2017 9:48AM - 10:00AM |
K33.00008: Comptational Design Of Functional CA-S-H and Oxide Doped Alloy Systems Shizhong Yang, Lokeshwar Chilla, Yan Yang, Kuo Li, Scott Wicker, Guang-Lin Zhao, Ebrahim Khosravi, Shuju Bai, Boliang Zhang, Shengmin Guo Computer aided functional materials design accelerates the discovery of novel materials. This presentation will cover our recent research advance on the Ca-S-H system properties prediction and oxide doped high entropy alloy property simulation and experiment validation. Several recent developed computational materials design methods were utilized to the two systems physical and chemical properties prediction. A comparison of simulation results to the corresponding experiment data will be introduced. [Preview Abstract] |
Wednesday, March 15, 2017 10:00AM - 10:12AM |
K33.00009: First Principles Calculation of Formation Energy for Nd-Fe-B compounds Adie Hanindriyo, Soumya Sridar, K.C. Hari Kumar, Ryo Maezono The compound Nd$_{2}$Fe$_{14}$B is widely used in manufacturing permanent magnets. This interest has driven much of the research on the Nd-Fe-B system including the computational calculation of phase diagrams (CALPHAD). Formation energy of Nd-Fe-B compounds is required for CALPHAD input, and are calculated using Density Functional Theory (DFT). Hubbard U correction is used to account for the localized nature of Nd 4f electrons. Comparison between 2 methods of choosing Hubbard U value is drawn: a simplified implementation of DFT+U using effective Hubbard U value (U$_{eff}$ = U-J) and from constrained random phase approximation (cRPA) for bulk Nd. It is shown that the former method is insufficient to accurately calculate formation energy of several Nd-Fe-B compounds. Comparison is also drawn between 2 types of pseudopotentials, norm-conserving and ultrasoft pseudopotentials. [Preview Abstract] |
Wednesday, March 15, 2017 10:12AM - 10:24AM |
K33.00010: Multi-level approach for design and synthesis of energetic materials with tailored properties Philip Pagoria, Roman Tsyshevskiy, Guzel Garifzianova, Aleksandr Smirnov, Maija Kuklja A typical approach to design novel high energy density materials usually involves sophisticated synthesis procedures combined with extensive sensitivity characterization tests. Such empirical explorations are time and effort consuming and often very expensive while the successful outcomes are never guaranteed. Thus, many new energetic materials were attained only to be rejected due to their dangerously high sensitivity. Therefore, an efficient approach that could potentially replace rather expensive trial and error method has yet to be established. Here we discuss a multi-level protocol developed for searching for novel energetic materials with tailored properties. Our approach combines state-of-the-art first principles modeling and semi-empirical big data analysis with advanced synthetic procedures and experimental characterization. We demonstrate how this protocol has been used to design fused and linear heterocyclic materials. [Preview Abstract] |
Wednesday, March 15, 2017 10:24AM - 10:36AM |
K33.00011: High-resolution functional renormalization group calculations for interacting fermions on the square lattice Julian Lichtenstein, David S\'anchez de la Pe\~na, Daniel Rohe, Edoardo Di Napoli, Carsten Honerkamp We derive a novel computational scheme for functional Renormalization Group (fRG) calculations for interacting fermions on 2D lattices [1]. The scheme is based on the exchange parametrization fRG for the two-fermion interaction, with additional insertions of truncated partitions of unity. These insertions decouple the fermionic propagators from the exchange propagators and lead to a separation of the underlying equations. We demonstrate that this separation is numerically advantageous and may pave the way for refined, large-scale computational investigations. Furthermore, on the basis of speedup data gained from our implementation, it is shown that this new variant facilitates efficient calculations on a large number of multi-core CPUs. We apply the scheme to the $t$,$t'$ Hubbard model on a square lattice to analyze the convergence of the results with the bond length of the truncation of the partition of unity. Due to the computational performance of the implementation a high resolution in momentum space can be achieved, which allows for an analysis of long ranged interactions.\\ \ [1] J. Lichtenstein, D. S\'{a}nchez de la Pe\~{n}a, et al., \textit{ArXiv e-prints} (2016), arXiv:1604.06296. [Preview Abstract] |
Wednesday, March 15, 2017 10:36AM - 10:48AM |
K33.00012: Entanglement entropy and computational complexity of the Anderson impurity model out of equilibrium Zhuoran He, Andrew Millis We study the growth of entanglement entropy in the real-time dynamics of the Anderson impurity model, with particular focus on the quenched single-impurity Anderson model (SIAM) out of equilibrium. A class of polynomial-time solvable models by the density matrix renormalization group (DMRG) method are identified, in which the entanglement entropy at the maximum entropy cut of the bath grows logarithmically with time. The logarithmic growth of entropy is numerically found to be independent of the Hubbard $U$ in nonequilibrium SIAM. An energy criterion for such polynomial-time complexity is proposed and solvable cases are also found in noninteracting multi-impurity models and periodically driven models. [Preview Abstract] |
Wednesday, March 15, 2017 10:48AM - 11:00AM |
K33.00013: Nanoscale Charge Balancing Mechanism in Calcium-Silicate-Hydrate Gels: Novel Complex Disordered Materials from First-principles Ongun Ozcelik, Claire White Alkali-activated materials which have augmented chemical compositions as compared to ordinary Portland cement are sustainable technologies that have the potential to lower CO$_2$ emissions associated with the construction industry. In particular, calcium-silicate-hydrate (C-S-H) gel is altered at the atomic scale due to changes in its chemical composition. Here, based on first-principles calculations, we predict a charge balancing mechanism[1] at the molecular level in C-S-H gels when alkali atoms are introduced into their structure. This charge balancing process is responsible for the formation of novel structures which possess superior mechanical properties compared to their charge unbalanced counterparts. Different structural representations are obtained depending on the level of substitution and the degree of charge balancing incorporated in the structures. The impact of these charge balancing effects on the structures is assessed by analyzing their formation energies, local bonding environments, diffusion barriers and mechanical properties. These results provide information on the phase stability of alkali/aluminum containing C-S-H gels, shedding light on the fundamental mechanisms that play a crucial role in these complex disordered materials. [1] arXiv:1610.00112 [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