Bulletin of the American Physical Society
APS March Meeting 2022
Volume 67, Number 3
Monday–Friday, March 14–18, 2022; Chicago
Session W32: Material Science and AIRecordings Available
|
Hide Abstracts |
Sponsoring Units: GDS Chair: Gabriel Landi, Instituto de F??sica da Universidade de Room: McCormick Place W-192B |
Thursday, March 17, 2022 3:00PM - 3:12PM |
W32.00001: Machine-Learning Assisted First-Principles Model Development to Interpret ARPES Data and Assist Inverse Design of Heterostructures Sanghamitra Neogi, Artem Pimachev Advances of nanofabrication techniques have achieved great control over the growth of semiconductor heterostructures. Nevertheless, fabrication of heterostructures is strongly affected by strain environent in component layers, and the resulting electronic properties show high variability. The layer compositions and external substrate induced strains prompt non-uniform separations between monolayers and modulate electronic properties. It remains a challenge to model electronic transport coefficients of technologically relevant heterostructures incorporatingfull structural complexity, representing the vast fabrication dependent structural parameterspace. On the other hand, characterization techniques like the angle-resolved photoemission spectroscopy (ARPES) provides great insight regarding the nature of electronic bands. We establish a convolutional neural network based autoencoder that can extract infomation from ARPES images and use first principles data to predict electronic properties of fabricated heterostructures. |
Thursday, March 17, 2022 3:12PM - 3:24PM |
W32.00002: Structure determination from theory and experiments Venkata Surya Chaitanya Kolluru, Eric Schwenker, Davis G Unruh, Maria K Chan The atomistic structure determines the stability and properties of a material and its potential applications. We develop the software tools Ingrained and FANTASTX (Fully Automated Nanoscale To Atomistic Structure from Theory and eXperiments) to determine atomistic structures from experimental data with the help of theory and machine learning. The ingrained software [1] constructs a grain boundary structure or validates a surface structure based on experimental STEM or STM images, respectively. We will show examples of grain boundary structures created using Ingrained which are used as starting points for further analysis. This provides a path to understand complex mechanisms from characterization data. We demonstrate the utility of Ingrained which was used to successfully determine the surface structure of hydrogenated borophene [2]. FANTASTX is a multi-objective genetic algorithm tool that helps to find the thermodynamically or kinetically stabilized structures observed experimentally. We will discuss the use of FANTASTX to determine interfacial and grain boundary atomistic structures from DFT and STEM images. |
Thursday, March 17, 2022 3:24PM - 3:36PM |
W32.00003: Artificial intelligence guided studies of two-dimensional Ising ferromagnets Trevor D Rhone, Vaishnavi D Neema, Daniel T Larson, Efthimios Kaxiras The discovery of van der Waals (vdW) materials with intrinsic magnetic order in 2017 has given rise to new avenues for the study of emergent phenomena in two dimensions. In particular, a monolayer of CrI3 was found to be an Ising ferromagnet. Other vdW transition metal halides, such as CrBr3, were later found to have different magnetic properties. How many vdW magnetic materials exist in nature? What are their magnetic properties? How do these properties change with the number of layers? A conservative estimate for the number of candidate vdW materials (including monolayers, bilayers and trilayers) exceeds ~106. A recent study showed that machine learning can be exploited to discover new vdW Heisenberg ferromagnets based on Cr2Ge2Te6 [1]. In this talk, we will use materials informatics – materials science combined with artificial intelligence (AI) – as a tool to efficiently explore the large chemical space of vdW transition metal halides and to guide the discovery of magnetic vdW materials with desirable spin properties. That is, we investigate crystal structures based on monolayer Cr2I6 of the form A2X6, which are studied using density functional theory (DFT) calculations and AI. Magnetic properties, such as the magnetic moment are determined. The formation energy is also calculated and used as a proxy for the chemical stability. We show that AI, combined with DFT, can provide a computationally efficient means to predict properties of vdW magnets. In addition, data analytics provides insights into the microscopic origins of magnetic ordering in two dimensions. We also explore how our study of magnetic monolayers can be extended, with proper modification, to multilayer vdW materials. This non-traditional approach to materials research paves the way for the rapid discovery of chemically stable magnetic vdW materials with potential applications in spintronics and data storage. |
Thursday, March 17, 2022 3:36PM - 3:48PM Withdrawn |
W32.00004: Autonomous materials synthesis via hierarchical active learning of non-equilibrium phase diagrams Maximilian Amsler, Sebastian Ament, Duncan Sutherland, Ming-Chiang Chang, Dan Guevarra, Aine Connolly, John M Gregoire, Michael O Thompson, Bruce van Dover, Carla P Gomes Autonomous experimentation enabled by artificial intelligence (AI) offers a new paradigm for accelerating scientific discovery. Non-equilibrium materials synthesis is emblematic of complex, resource-intensive experimentation whose acceleration would be a watershed for materials discovery and development. The mapping of non-equilibrium synthesis phase diagrams has recently been accelerated via high throughput experimentation but still limits materials research because the parameter space is too vast to be exhaustively explored. We demonstrate accelerated synthesis and exploration of metastable materials through hierarchical autonomous experimentation governed by the Scientific Autonomous Reasoning Agent (SARA). SARA integrates robotic materials synthesis and characterization along with a hierarchy of AI methods that efficiently reveal the structure of processing phase diagrams. SARA designs lateral gradient laser spike annealing (lg-LSA) experiments for parallel materials synthesis and employs optical spectroscopy to rapidly identify phase transitions. Efficient exploration of the multi-dimensional parameter space is achieved with nested active learning (AL) cycles built upon advanced machine learning models that incorporate the underlying physics of the experiments as well as end-to-end uncertainty quantification. With this, and the coordination of AL at multiple scales, SARA embodies AI harnessing of complex scientific tasks. We demonstrate its performance by autonomously mapping synthesis phase boundaries for the Bi2O3 system, leading to orders-of-magnitude acceleration in establishment of a synthesis phase diagram that includes conditions for kinetically stabilizing δ-Bi2O3 at room temperature, a critical development for electrochemical technologies such as solid oxide fuel cells. |
Thursday, March 17, 2022 3:48PM - 4:00PM |
W32.00005: Inverse design of two-dimensional materials with invertible neural networks Victor Fung, Jiaxin Zhang, Guoxiang Hu, Panchapakesan Ganesh, Bobby G Sumpter The ability to readily design novel materials with chosen functional properties on-demand represents a next frontier in materials discovery. However, thoroughly and efficiently sampling the entire design space in a computationally tractable manner remains a highly challenging task. To tackle this problem, we propose an inverse design framework (MatDesINNe) utilizing invertible neural networks which can map both forward and reverse processes between the design space and target property. This approach can be used to generate materials candidates for a designated property, thereby satisfying the highly sought-after goal of inverse design. We then apply this framework to the task of band gap engineering in two-dimensional materials, starting with MoS2. Within the design space encompassing six degrees of freedom in applied tensile, compressive and shear strain plus an external electric field, we show the framework can generate novel, high fidelity, and diverse candidates with near-chemical accuracy. We extend this generative capability further to provide insights regarding metal-insulator transition, important for memristive neuromorphic applications among others, in MoS2 which is not otherwise possible with brute force screening. This approach is general and can be directly extended to other materials and their corresponding design spaces and target properties. |
Thursday, March 17, 2022 4:00PM - 4:12PM |
W32.00006: Learning design rules for selective oxidation catalysts from high-throughput experimentation and artificial intelligence Lucas Foppa, Christopher A Sutton, Luca M Ghiringhelli, Sandip De, Patricia Löser, Stephan Schunk, Ansgar Schäfer, Matthias Scheffler The design of heterogeneous catalysts is challenged by the complexity of materials and processes that govern reactivity and by the fact that the number of good catalysts is very small. Here, we show how the subgroup-discovery (SGD) artificial-intelligence local approach[1] can be applied to an experimental plus theoretical data set to identify constraints or rules on key physicochemical parameters that exclusively describe materials and reaction conditions with outstanding catalytic performance.[2] By using high-throughput experimentation, 120 SiO2-supported catalysts containing Ru, W and P were synthesized and tested in propylene oxidation. As candidate descriptive parameters, the temperature and ten calculated parameters related to the composition and chemical nature of elements in the catalyst materials, were offered. The temperature, the P content, and the composition-weighted electronegativity are identified as key parameters describing high yields of value-added oxygenate products. The SG rules reflect the underlying processes associated to high performance, and guide catalyst design. |
Thursday, March 17, 2022 4:12PM - 4:24PM |
W32.00007: CCSS-2D: Computational Catalogue of Spin Splittings in 2D Elton Ogoshi de Melo, Gabriel M Nascimento, Carlos Mera Acosta, Gustavo M Dalpian Materials databases have shown to be fundamental to workflows in Materials Informatics, serving as a starting point for the search of new materials and new physical insights. With the intent of contributing to the growing ecosystem of high-throughput databases of DFT calculations, here we provide a highly structured catalogue of spin splitting (SS) properties of 2D materials. SS effects, resulting from breaking of spin degeneracy, are the main target functionalities in spintronics applications. Identification and design of SS prototypes (Rashba, Dresselhaus and Zeeman) are fundamental, and mechanisms for the SS control are also indispensable for applications in devices. The growing number of proposed and fabricated 2D materials may offer novel routes for SS control. Here we used an inverse design process, based on the enabling design principles of SS, to select 437 materials from the C2DB database. DFT calculations with spin-orbit coupling were carried out to identify and classify them according to their SSs. The result is an extensive and descriptive catalogue of SSs, from which we employed a Bayesian analysis to investigate chemical and structural trends, contributing to a rationalized design of 2D materials and assessing the potential usage of the new generated data. |
Thursday, March 17, 2022 4:24PM - 4:36PM |
W32.00008: Symbolic Regression for Materials Science: A Case Study for Lattice Thermal Conductivity Thomas A Purcell, Matthias Scheffler, Luca M Ghiringhelli, Christian Carbogno Artificial intelligence (AI) frameworks that are capable of creating reliable and interpretable models are paramount for discovering new functional materials. Here we present the SISSO++ code [2], a new implementation of the sure independence screening and sparsifying operator (SISSO) approach [1]. As a combination of compressed sensing and symbolic regression, SISSO can deterministically find the optimal analytic expression for a property, to a user-defined level of complexity. SISSO++ provides both a high performance library to perform SISSO calculations and a user-friendly python interface that facilitate interfacing SISSO into existing frameworks. SISSO++ also includes updates to the SISSO methodology itself, e.g. non-linear parameterization of features and using multiple residuals for each iteration, by quantitatively developing an analytical representation of the thermal conductivity. In particular, we discuss the link between a material's anharmonicity [3] and its thermal conductivity and demonstrate how the developed formalism accelerates rational materials design and discovery by orders of magnitude. |
Thursday, March 17, 2022 4:36PM - 4:48PM |
W32.00009: Autoencoder-based feature space of XRD peak patterns Kenta Hongo, Ryo Maezono, Kousuke Nakano, Keishu Utimula Recent advances in experiments enable us to obtain high-throughput X-ray diffraction (XRD) data, but their characterization is still far from automation. This is because there is no procedure for automatically identifying relevant peaks to characterize the corresponding structure, i.e., either all or specific peaks are necessary for the characterization. To address this, we establish an autoencoder-based scheme for identifying the peak relevance. We first construct an autoencoder neural network so as to map each of XRD patterns onto each of 2-dimensional points in the feature space. This encoder is then used to test the peak relevance: For a given XRD pattern, we mask its concerned peak and input the masked XRD pattern into the encoder. If the masked point significantly shifts from the original one in the feature space, then the masked peak can be considered relevant for characterizing the XRD pattern. Interestingly, applying this scheme to magnetic alloys, we found a low relevant peak having a significant intensity, which is useful for characterizing the XRD pattern. This finding was heuristically revealed by machine learning, which cannot be easily interpreted from physical viewpoints such as higher-order reflections etc. |
Thursday, March 17, 2022 4:48PM - 5:00PM |
W32.00010: Discovery of New Air Separation Metal-Organic Frameworks via Bayesian Optimization Eric Taw, Jeffrey B Neaton The discovery of new metal-organic frameworks (MOFs) for gas separations is highly labor- and time-intensive. Recent studies propose using high-throughput computational screening, though this approach requires a trade-off between computational expense and exploration of a wide chemical space. Because density functional theory (DFT) calculations are expensive and often require manual intervention, DFT is often limited to a small set of structurally similar MOFs. Here, we mitigate the expense-exploration trade-off by using Bayesian optimization to minimize the number of expensive yet accurate DFT calculations required to evaluate a new MOF for air separation. We first benchmark the ability for PBE-D3+U calculations to recover experimentally determined enthalpies of adsorption of O2, then proceed to calculate the same for promising candidate MOFs proposed by our Bayesian optimization procedure. By identifying MOFs that have a binding enthalpy of O2 around -45 kJ/mol, we propose a candidate material that is predicted to dramatically reduce the energy cost of air separation compared to standard industrial adsorbents. |
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. |
© 2025 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