Computational Discovery and Design of Emerging Materials: Session IV
Sponsored by: TMS Materials Processing and Manufacturing Division, TMS: Computational Materials Science and Engineering Committee
Program Organizers: Arunima Singh, Arizona State University; Houlong Zhuang, Arizona State University; Sugata Chowdhury, National Institute of Standards and Technology; Arun Kumar Mannodi Kanakkithodi, Purdue University
Tuesday 2:00 PM
February 25, 2020
Room: 32B
Location: San Diego Convention Ctr
Session Chair: Arunima Singh, Arizona State University
2:00 PM Invited
Predicting Functional Defects by Design in Energy and Quantum Materials: Panchapakesan Ganesh1; 1Oak Ridge National Laboratory
Defects and interfaces determine and control properties of solid-state materials and to a large degree impart specific functionalities. E.g. kinetics of ionic transport fundamentally limits the performance, efficiency and operating conditions of almost all known renewable energy technologies, ranging from Li- ion batteries to fuel-cells in Energy storage & conversion materials; Creation and motion of defects can trigger concomitant ‘switchable’ metal-insulator, structural and magnetic- phase- transitions showing coupled hysterics for Neuromorphic computing; Defects and interfaces can be used to modify topology of a solid leading to emergent new quantum-phases (say) for dissipationless, low-power quantum-transport, in topological quantum materials or can play the ugly role of quenching emergent topological properties. The challenge lies both in identifying relevant defects, accurately capturing its fundamental nature and influence on material functionality and using this knowledge to design improved materials, using state-of-the-art computational materials approaches, integrated with complimentary data-analytics as well as experimental capabilities. In this talk, I will attempt to highlight some of our recent as well as ongoing activities in addressing this challenge.
2:30 PM
The Relationship Between Compositional Mixing and Phase Stability of Metal-halide Perovskites: Theoretical Study: Ki-Ha Hong1; 1Hanbat National Univeristy/Department of Materials Science and Engineering
Metal-halide perovskites (MHPs) are promising candidates for next-generation photovoltaic and photoelectronic materials. Polymorphism control between perovskite and non-perovskite structures is one of the critical issues in highly stable and efficient MHP-based applications. Compositional mixing is one of the efficient ways to overcome this phase-selection issue. The roles of compositional mixing on the phase stability are investigated with density functional theory (DFT). The concept of the ‘anti-stabilizer’ for the proper selection of the preferred phase is suggested by our previous study. We present that the anti-stabilizing concept can be successfully applied to stabilize the proper band gap polymorph of black CsSnI3 and Bi-, and Sb-based perovskites. Additionally, the roles of additives including small alkali metal interstitials are discussed concerning thermodynamics.
2:50 PM
First-principles-based Hybrid Perovskite Materials Design for Memristor: Donghwa Lee1; Seong Hun Kim1; 1Pohang University of Science and Technology (POSTECH)
MAPbI3-based memristors have attracted many researchers since it shows various advantages such as forming-free operation, a low operating voltage, high on/off ratio. However, low endurance and retention time needs to be overcome for its practical application. Thus, there are pressing need to develop alternative perovskite materials. However, the combinations of organic molecules and inorganic constituents are too numerous to investigate all possible candidate materials experimentally. In this study, therefore, a computational approach is used to find out alternative perovskite materials that can replace conventional MAPbI3-based hybrid perovskite materials. We have extensively studied numerous candidate materials by incorporating different transition metal elements or organic molecules into various backbone structures to find out new perovskite materials. Our study has identified a couple of candidates that can possibly apply for the resistance switching layer. In addition, analysis on electronic structure has led the understanding of the relationship between constituent elements and properties.
3:10 PM
Discovery and Characterization of 1D Inorganic Polymers Through Datamining and Density Functional Theory: Joshua Paul1; Janet Lu1; Sohum Shah1; Stephen Xie1; Richard Hennig1; 1University of Florida
Two-dimensional (2D) materials have been a topic of significant study since the discovery of free-standing graphene in 2004. One-dimensional (1D) materials are a much less studied class of sub-periodic materials but are not lacking in interesting phenomena. We search the MaterialsProject database for bulk materials composed of aligned 1D chains and isolate them, identifying several hundred inorganic polymers. We then focus our calculations to those with d and f-valence elements and calculate the stability, electronic, and magnetic characteristics of the systems. We compare the stability of the chains calculated to those that have been synthesized to propose an initial standard for determining metastability of 1D materials. Following, we identify a range of electronic behavior in these chains, including half-metals, band gaps greater than 3 eV, and magnetic insulators with differing band gaps greater than 2 eV. We then identify potential challenges in preparing these materials and characterizing their magnetic behavior.
3:30 PM Break
3:50 PM Invited
Toward Rational Design and Discovery of Metastable Materials: Vladan Strevanovic1; 1Colorado School Of Mines
Metastable solids, such as diamond, glass or solid chocolate, are invaluable in our daily lives. However, despite a rather extensive knowledge of metastability our ability to rationally discover and design metastable forms of matter is rather limited. In this talk I will discuss realizability of metastable crystalline phases (polymorphs) as related to the volumes of configuration space occupied by the local minima on the potential energy surface; and how an alternative description of glassy solids can be developed from the statistical treatment of crystalline local minima obtained from the first-principles random structure search. I will also describe key results from our recent work on large-scale predictions of the kinetics of polymorphic transformations. In all of these areas our recent developments offer a path to quantitative predictions of relevant properties without experimental inputs, which opens the door to fully predictive description and future design of metastable materials for various applications.
4:20 PM
Predicting the Properties of Crystals with High Accuracy Using Deep Learning: Weike Ye1; Chi Chen1; Zhenbin Wang1; Iek-Heng Chu1; Yunxing Zuo1; Chen Zheng1; Shyue Ping Ong1; 1University of California San Diego
Being able to predict the properties of crystals is a prerequisite to materials discovery. First-principle approaches, such as density functional theory (DFT), are expensive and scale poorly with system size. Here, we demonstrate that machine learning (ML) can provide a shortcut. We show that neural networks (NN) with only two descriptors, the electronegativity and ionic radius, can predict crystal stability within DFT error. The mean absolute errors of the NN-predicted DFT formation energies for two sample system types, garnets and perovskites, are only 7-13 meV/atom and 30-34 meV/atom, respectively. We also developed universal MatErials Graph Network (MEGNet) models that can predict properties for both molecules and crystals. MEGNet models provide highest accuracy to-date in predicting 11 of 13 properties of the molecules in QM9, and 4 properties of crystals in Materials Project. Furthermore, we provide successful approaches to often-encountered problems in materials informatics such as disordered structures and data limitation.
4:40 PM
Introducing the MEAM Interatomic Potential for NiTiHf Shape Memory Alloys: Meghnath Jaishi1; Garitt Tucker1; Aaron Stebner1; 1Colorado School of Mines
NiTiHf based shape memory alloys (SMAs) have shown considerable promise in recent days as a viable alternative to their parent NiTi (nitinol) SMAs. Because of their high strength, and wide variability in martensitic transformation temperature ranging from below room temperature to above 100 ēC enable their usages from biomedical applications to aviation industries. However, the lack of atomistic understanding during their microstructural engineering by ageing, alloying and processing techniques is inhibiting these smart materials from its wide scale applications. Atomistic modeling to unlock the atomic level information for an efficient shape memory application requires an interatomic potential with a high predictive accuracy. Herein, we introduce the optimized interaction potential for NiTiHf SMAs within the framework of modified embedded atom method (MEAM). Testing of various mechanical and microstructural features obtained from experimental measurements as well as ab-initio calculations proves the highly accurate predictive nature of our interatomic potential for NiTiHf SMAs.
5:00 PM
Density Functional Theory and Machine Learning Guided Prediction of Thermal Properties of Rare-earth Disilicates: Mukil Ayyasamy1; Prasanna Balachandran1; 1University of Virginia
Application of rare-earth disilicates in extreme environments, such as the environmental barrier coatings (EBC) in jet engines, require knowledge of their coefficient of thermal expansion (CTE). Till date, CTE data have been determined experimentally for only a small fraction of the vast search space. The present work explores a computational approach, enabled by density functional theory (DFT) calculations and machine learning (ML) methods, to accelerate the prediction of CTE in previously unexplored compounds. DFT calculations were performed to construct the descriptors for ML. Ensemble ML models were used to establish a relationship between the descriptors from DFT and CTE (taken from published experiments). The trained models were then used to rapidly predict the CTE for previously unexplored compounds. One of the key outcomes is the identification of novel disilicates with CTE in the range (3-5.5×10-6 K-1) that are suitable for EBC applications. These compounds are recommended for experimental validation.