Computational Discovery and Design of Emerging Materials: Session VI
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

Thursday 2:00 PM
February 27, 2020
Room: 32B
Location: San Diego Convention Ctr

Session Chair: Houlong Zhuang, Arizona State University


2:00 PM  Invited
Electronic Excitations and Ultrafast Dynamics: Pushing Towards Materials Engineering and Design: Andre Schleife1; 1University of Illinois at Urbana-Champaign
    Excited electronic states and their dynamics underlie how we use and probe matter. Recent experimental advances allow doing so with unprecedented accuracy and time resolution, however, interpretation relies on theoretical insight e.g. from first-principles theoretical-spectroscopy. In this talk, I will illustrate recent examples for how many-body perturbation theory and time-dependent density functional theory successfully lead to deep understanding of light absorption of organo-metal halides and enhancement of defect diffusion by hot electrons due to particle radiation. While these first-principles techniques allow for predictive accuracy and excellent agreement with experiment, they rely on approximations, and I will illustrate our recent efforts of developing better understanding of dielectric screening physics. I will also describe how we bridge time scales from ultrafast electron dynamics to atomic diffusion. Finally, I will describe how incorporating online databases into computational research can side-step the problem of high computational cost to facilitate electronic materials design.

2:30 PM  
Analysis of Chemical Activity of Bismuthene in the Presence of Environment Gas Molecules by Means of Ab-initio Calculations: Elena Korznikova1; Andrey Kistanov1; Salavat Khadiullin2; 1Russian Academy of Sciences; 2Ufa State Aviation Technical University
    Recently a new group of two-dimensional (2D) materials, originated from the group V elements (pnictogens), has been successfully synthesized and has gained the global attention owing to its good structural integrity and various outstanding properties. Due to the high surface-volume ratio and extraordinary chemical activity, 2D pnictogens are highly sensitive to the exposure of the environment and external adsorbates. This work perform a systematical first-principles investigation on the effects of the environmental oxygen and water molecules on the structural stability of newly emergened group V 2D material bismuthene. The interaction of oxygen molecules with bismuthene is found to be much stronger than that of water molecules. The present work uncovers the oxidation mechanisms and suggests the ways for maintaining the stability of bismuthene and its 2D pnictogens counterparts, which may be important for their fabrication, storage, and applications. Work was supported by Russian Foundation for Basic Research, grant No 18-32-20158.

2:50 PM  
Accelerating the Genetic Algorithm for Structure Prediction in 2D Materials using Machine Learning: Stephen Xie1; Shreyas Honrao2; Anne Marie Tan1; Richard Hennig1; 1University of Florida Department of Materials Science and Engineering; 2NASA Ames Research Center
    We present our machine learning approach to accelerating global structure prediction by coupling the Genetic Algorithm for Structure Prediction (GASP) to surrogate machine learning models for energy prediction. Using a small number of structurally-diverse materials generated with GASP and their formation energies from density functional theory, we train interatomic potentials using support vector regression. We show that such potentials can be used to filter low-value candidates, reducing the computational cost of the genetic algorithm by eliminating materials with a high probability of having higher energy. As more materials are generated and evaluated, their inclusion in the training data iteratively improves the accuracy of the surrogate model. We discuss the tuning of radial and angular distribution functions to encode relevant physical information into machine-readable inputs. Furthermore, we demonstrate how augmenting the training data with local energies and forces improves model performance. Finally, we apply our approach to two-dimensional group-III chalcogenide systems.

3:10 PM  
Tunability of Martensitic Transformation in Mg-Sc Shape Memory Alloys: a DFT Study: Shivam Tripathi1; Karthik Guda Vishnu1; Michael Titus1; Alejandro Strachan1; 1Purdue Univeristy
    Low density makes Mg-Sc shape memory alloys attractive for a wide range of applications, unfortunately use of these alloys is hindered by low martensitic transformation temperature(173K). We use DFT to characterize the energetics associated with the martensitic transformation in a Mg-Sc(19.44at.% Sc) alloy from a disordered BCC austenite to a disordered orthorhombic martensite. The simulations predict lattice parameters and XRD in good agreement with experiments and the martensite to be 10.5 meV/atom lower in energy than austenite at zero temperature. We also explore the effect of epitaxial strain with the objective of increasing the transformation temperature. Bi-axial strain between 5 and 7% increases the zero-temperature energy difference between the phases by over 60%. Similar stabilization of the martensite can be achieved by the addition of pure Mg as a coherent second phase. These results indicate that coherency strains can be used to increase the martensitic transformation temperature to room temperature.

3:30 PM Break

3:50 PM  Invited
Computational Discovery of Strongly Correlated Quantum Matter through Downfolding: Hitesh Changlani1; 1Florida State University
    Due to advances in computer hardware and new algorithms, it is now possible to perform highly accurate many-body simulations of realistic materials with all their intrinsic complications. The success of these simulations leaves us with a conundrum: how do we extract useful physical models from these simulations? There is a clear need for a multi-scale approach that can "downfold" strongly correlated materials to model Hamiltonians such as the Hubbard, Heisenberg and Kitaev models, and then accurately solve this model. Knowledge of the effective Hamiltonian should help accelerate the discovery of new phases of matter, such as quantum spin liquids, in addition to contributing to predictive theories of strongly correlated matter. My talk will present progress on a formal theory of "density matrix downfolding" and also showcase work (done recently in collaboration with neutron scattering experimentalists) showing the possible existence of a S=1 spin liquid in a nickel based pyrochlore magnet.

4:20 PM  
Influence of Strain on Mesoscopic 2D Film Growth from Phase Field Methods: Tara Boland1; Arunima Singh1; 1Arizona State University
    The emergence of two dimensional materials opened up many potential avenues for novel device applications such as nanoelectronics, topological insulators, field effect transistors, microwave and terahertz photonics and many more. The incorporation of this technology into device applications however has been hindered due to the difficulty synthesizing and stabilizing the 2D materials. Traditional methods such as chemical vapor deposition result in films with defects and grain boundaries. Controlling the growth of these films requires a systematic understanding of the crucial factors of the film-substrate adhesion strength and mismatch strain. We present density functional theory calculations with van der Waals corrections and phase field methods to inform on the atomistic as well as mesoscopic growth mechanisms of 2D materials on various substrates.

4:40 PM  
Predicting Polymer Crystallinity Using Multi-fidelity Information Fusion with Machine Learning: Shruti Venkatram1; Lihua Chen1; Rampi Ramprasad1; 1Georgia Institute of Technology
     As renewable energy sources are increasingly gaining traction as an alternative to scarce fossil fuels, exploratory research on energy storage devices, particularly Li-ion batteries (LIBs), has become prolific. To safely utilize a high-performing LIB, a few alternatives have been suggested, one of which is replacing the liquid electrolyte with a suitable solid polymer electrolyte (SPE). SPEs offer several advantages over conventional liquid electrolytes, viz., low flammability, good processability, and no leakage issues. SPEs also eliminate the need for a separator, thereby decreasing the chances of an accident. A promising SPE candidate should have high Li-ion conductivity, a low glass transition temperature - both of which depend on the degree of crystallinity of the polymer.Data-driven efforts to predict the polymer crystallinity have been scarce. In this first-of-its-kind work, we develop a multi-fidelity dataset of over 400 polymers which comprises of a high-fidelity dataset which uses explicit experimental crystallinity information and another low-fidelity dataset which uses theoretical group contribution methods with experimental data. With this dataset, we then develop and compare machine learning models viz. conventional gaussian process regression with the high-fidelity dataset and co-kriging with the multi-fidelity dataset. Through this effort we aim to predict the polymer crystallinity of new polymers instantly and also use it as a screening criterion to design new materials for solid polymer electrolytes.