Computational Thermodynamics and Kinetics: On-Demand Oral Presentations
Sponsored by: TMS Functional Materials Division, TMS Materials Processing and Manufacturing Division, TMS Structural Materials Division, TMS: Chemistry and Physics of Materials Committee
Program Organizers: Vahid Attari, Texas A&M University; Sara Kadkhodaei, University Of Illinois Chicago; Eva Zarkadoula, Oak Ridge National Laboratory; Damien Tourret, IMDEA Materials Institute; James Morris, Ames Laboratory
Monday 8:00 AM
March 14, 2022
Room: Physical Metallurgy
Location: On-Demand Room
Magnetic-field-induced Phonon Stiffening Enhances Demagnetization Entropy in a Shape Memory Alloy: Michael Manley1; Paul Stonaha1; Nick Bruno2; Ibrahim Karaman2; Raymundo Arroyave2; Douglas Abernathy1; Songxue Chi1; Jeffrey Lynn3; 1Oak Ridge National Laboratory; 2Texas A&M University; 3NIST
Magnetocaloric materials are of considerable interest due to their application in technologies such magnetic refrigeration. A class of magnetocaloric materials are magnetic shape memory alloys, which undergo a first-order phase transition at their martensitic transformation temperature. This talk will present a neutron scattering study of the lattice dynamics of the metamagnetic shape memory alloy Ni45Co5Mn36.6In13.4. We identify an anomalous temperature-dependent softening of an optic phonon below the Curie temperature, which we attribute to magneto elastic coupling. Further, we use in situ application of a magnetic field to show that the phonon is indeed coupled to the magnetism in the sample. Finally, we use measurements of the phonon density of states to show that the magnetic-field-induced phonon stiffening significantly enhances the demagnetization entropy.
Modelling Columnar-to-equiaxed Transition during Additive Manufacturing: Durga Ananthanarayanan1; Niklas Holländer Pettersson1; Greta Lindwall1; 1Kth Royal Institute Of Technology
Columnar-to-equiaxed transition during directional solidification is typically modelled analytically taking into account different factors such as dendrite growth kinetics, constitutional supercooling, and number of nucleating sites for different thermal gradients and solidification rates. In additive manufacturing processing conditions, solute trapping also becomes important at higher solidification rates. In this work, we formulate a model to simulate columnar-to-equiaxed transition during additive manufacturing taking into account solute trapping and multicomponent interactions, which are important for non-dilute technical alloys. We apply this model to study columnar-to-equiaxed transition in different additively manufactured alloys where columnar-to-equiaxed transition is enabled through the addition of heterogeneous nucleating sites or inoculants. Through coupling with multicomponent thermodynamic databases, the importance of including rapid solidification effects and multicomponent interactions is highlighted. Finally, the simulation results are also compared with experimental measurements of the distributions of columnar and equiaxed grains in as-built microstructures.
Slow Interface Dynamics from Atomistic Simulations: Chad Sinclair1; Siavash Soltani1; Joerg Rottler1; 1Univ. of British Columbia
Understanding the processes that control the migration of phase interfaces and grain boundaries is complex owing to the fact that these processes are controlled by atomistic features at the boundary but occur on experimental timescales far from those achievable by conventional atomistic simulations. Attempts to obtain kinetics, particularly at low temperature or in the presence of diffusing solute, are particular challenging in this regard. Here we present a method for the temporal coarse graining of short atomistic simulations to access slow processes governing interface migration on long (much longer than achievable by MD simulation) timescales. We illustrate the method both for grain boundaries (at temperatures below the roughening temperature) and phase interfaces with solute present and show how this technique can bridge atomistic simulations and continuum models for interface mobility and energy.
Thermodynamics of Ferroelectrics beyond Phenomenological Landau Theory: A Case Study of PbTiO3: Zi-Kui Liu1; Jinglian Du1; Yi Wang1; Shunli Shang1; 1Pennsylvania State University
Thermodynamics of ferroelectric materials with characteristic ferroelectric to paraelectric (FE-PE) transitions has been described by the phenomenological Landau theory. Here we present a multiscale entropy approach that can predict the FE-PE transition from first-principles calculations based on density functional theory (DFT) without fitting parameters. The present approach considers the entropic contributions from the ferroelectric domain configurations in addition to electronic and quasiharmonic phonon contributions. With inputs exclusively from DFT-based first-principles calculations, the FE-PE transition temperature in PbTiO3 is predicted showing remarkable agreement with experimental observations. The present approach is supported by the observations that the macroscopically cubic paraelectric phase is microscopically tetragonal with polarization by x-ray absorption fine-structure structure technique and ab initio molecular dynamic simulations.
Machine Learning for Inverse Crystal Structure and Topology Design: Suvo Banik1; Troy Loeffler1; Rohit Batra2; Subramanian Sankaranarayanan1; Sukriti Manna1; 1University of Illinois-Chicago; 2Argonne National Lab
The most common and popular method for structure search and optimization are based on evolutionary design. This can often be cumbersome, limited to few tens of parameters and fails for large structural configurations or design problems with high degrees of freedom. Reinforcement learning approaches mostly operate in discrete action space such as in Go game but the applications of that to inverse problems is limited since most inverse problems deal with continuous action space. There are a large number of inverse structural search problems ranging from crystal structure search in material sciences to topology design in Quantum information, where it is highly desirable to optimize structure/configuration to target desired properties or functionalities. This talk will provide an overview of our current efforts to perform scalable crystal structure and topology search to discover and design metastable or non-equilibrium phases with desired functionality.
Uncertainty Driven Computational Thermodynamics: Noah Paulson1; Joshua Gabriel1; 1Argonne National Laboratory
In computational thermodynamics, statistics and uncertainty quantification are critical for evaluating the confidence in predictions and models with substantial potential for impact on materials development and qualification. The role for statistical techniques, especially Bayesian analysis, is not limited to obtaining uncertainty intervals on phase boundaries. These methods can drive the development of selection criteria for the best combinations of datasets and thermodynamic descriptions that support each other for a given material system. We present the results of such uncertainty driven investigations into thermodynamic model development for unary and binary metallic systems, as well as multi-component battery cathode materials. Specifically, we share statistical approaches to Bayesian model selection, parameter inference, uncertainty quantification/propagation, and the automated weighting of both experimental and computed datasets. This will be accompanied by a discussion of the use of statistical techniques in thermodynamics to inform critical design choices and gleaning scientific insights.
Phase Field Modelling of Microstructure Formation in Rapidly Solidified Steel: Nikolas Provatas1; Salvador Valtierra Rodriguez1; Damien Pinto1; Michael Greenwood2; 1McGill University; 2McMaster University
We present a multi-order phase-field model applied to the study solidification microstructure formation in rapidly solidified AISI309L austenitic stainless steel. We employ data from the CALPHAD literature to capture the thermodynamics of the phases present and use the Schaeffler diagram to express the Chromium and Nickel equivalent compositions. The thermal solidification history used to drive the system is based on a combination of thermocouple and FEM data generated for this study. We examine the spatial statistics of simulated dendritic microstructures and segregation and compare these with corresponding experimental data, finding good comparison. We conclude the talk with a recent machine learning study that uses simulation data to train a neural network algorithm to predict the time-evolution of microstructure and segregation during solidification. We show that this method can complement multi-scale algorithms to further enhance efficiency, or be used in stand-alone mode to predict microstructure evolution.
Phase-field Model of Solid Stoichiometric Compounds and Solution Phases: Yanzhou Ji1; Long-Qing Chen1; 1Penn State University
Phase-field modeling of stoichiometric phases is a long-standing issue since the chemical potential of a stoichiometric compound only exists at the stoichiometry points. Parabolic approximations to the chemical potential of stoichiometric compounds usually lead to thermodynamic inconsistency, difficulty in parameterization, and numerical problems. Here we develop a phase-field model for predicting microstructure evolution involving simultaneous solid stoichiometric and solution phases. We demonstrate its application using a well-known example of precipitation of stoichiometric è’ precipitates in a solid solution matrix. The proposed framework should be applicable to other common processes such as crystallization of stoichiometric compounds, vapor-phase deposition of stoichiometric thin films or two-dimensional materials, oxidation of alloys, electrochemical deposition, interfacial reactions, etc.
Data-driven Magnetic Materials Modeling; Advances in Classical Molecular Dynamics: Svetoslav Nikolov1; Mitchell Wood1; Aidan Thompson1; Julien Tranchida1; 1Sandia National Laboratories
We outline a data-driven simulation method for constructing machine learned interatomic potentials that accurately capture both magnetic and phononic degrees of freedom in iron. This novel approach allows us to incorporate realistic spin effects, typically only accessible in costly first-principles models, into computationally efficient classical molecular dynamics. To test our framework, we examine the magnetoelastic, magnetostrictive, and thermal properties of alpha-iron, which serves as an analogue for any ferromagnetic material. We find that our MD simulations capture the experimental trends in the magnon/phonon thermal conductivities and elastic properties well, up to and above the Curie temperature. In our conductivity analysis, we probe the dominant heat carrying modes within the magnon/phonon subsystems and investigate how these are modified with changes in temperature and external magnetic fields. Using our novel simulation method we also examine how the magneto-crystalline anisotropy energy and magnetostriction coefficients vary across the alpha-phase.
Vacancy Ordering in Zirconium Carbide Explored via First-principles Calculations and Calphad: Theresa Davey1; Ying Chen1; 1Tohoku University
Zirconium carbide (ZrCx) is an ultra-high temperature ceramic material with melting point around 3700 K and tuneable mechanical and thermodynamic properties thanks to its wide range of stable stoichiometry (~0.5 ≤ x ≤ 1) facilitated by varying numbers of carbon vacancies. Despite over a century of study, the low temperature equilibrium properties are relatively unknown due to high synthesis temperatures, significant impurity contamination, and slow vacancy diffusion preventing equilibration. Intermittent experimental investigations report vacancy ordering at around room temperature, which is consistent with several theoretical investigations. Short- and long-range ordering of carbon vacancies, as well as the disordered phase, was considered using first-principles calculations. The effects of temperature and oxygen impurities on vacancy ordering were examined. The results were incorporated into a Calphad database that describes the order-disorder transitions in the carbon-zirconium system, and is optimised considering all available experimental and calculated data.
Ab-initio Insights into the Impact of the Wet-synthesis Conditions on the Structure and Composition of Metal Nano-aerogels: Mira Todorova1; Su-Hyun Yoo1; Poulami Chakraborty1; Tilmann Hickel1; Se-Ho Kim1; Baptiste Gault1; Joerg Neugebauer1; 1Max-Planck-Insitut Fuer Eisenforschung
Metal nano-aerogels are a class of materials that combine high surface area, high structural stability, and superior catalytic activity towards a variety of chemical reactions. Their performance sensitively depends on their sub-nanoscale structure and the impurity distribution within the particles. Understanding how synthesis conditions affect the impurity integration and distribution opens routes to targeted design of aerogels with desired properties. Combining density functional theory calculations with thermodynamic considerations, we discuss how variations in the synthesis conditions impacts Pd nano-aerogels synthesized from a K2PdCl4 precursor and different concentrations of a NaBH4 reductant. We find that a delicate balance between thermodynamic and kinetic effects determines the concentration of Na and K impurities in the grain boundaries of the synthesized Pd nano-particles. Se-Ho Kim et al., submitted
Solute Drag Assessment of Grain Boundary Migration in Au Using Atomistic Simulations: Ayush Suhane1; Daniel Scheiber2; Maxim Popov2; Vsevolod Razumovskiy2; Lorenz Romaner3; Matthias Militzer1; 1The Centre for Metallurgical Process Engineering, The University of British Columbia, Vancouver, Canada; 2Materials Center Leoben Forschung GmbH, Roseggerstrasse 12, 8700 Leoben, Austria; 3Department of Materials Science, University of Leoben, Leoben, Austria
Solutes in metals and alloys can have a significant tendency for grain boundary (GB) segregation leading to reduced GB migration rates also known as solute drag. Here, the segregation energy and the solute diffusivity across the GB are important parameters that are typically adjusted to describe experimental observations. As an alternative we combine the Cahn-Lücke-Stüwe (CLS) solute drag model with density functional theory (DFT) simulations. To illustrate the proposed approach we analyze available experimental data for migration rates of the 30˚<111> GB in Au with Fe and Bi impurities at the ppm level. Based on DFT, Bi is identified as the most important segregating element. The Bi segregation and activation energies determined by DFT are within the uncertainties of the analysis consistent with those obtained by the conventional CLS solute drag fit. We critically discuss the strengths and quantitative limitations of the proposed approach.
Uncertainty Quantification for Ferromagnetic-Paramagnetic Phase Transition Onset by Integrating an Analytical Approach into Ising Models: Md Mahmudul Hasan1; Arulmurugan Senthilnathan1; Pinar Acar1; 1Virginia Tech
Ferromagnetic materials are widely used in magneto-mechanical devices such as sensors, motors, generators, and transformers. However, their magnetism reduces drastically when ferromagnetic-paramagnetic phase transition occurs at the critical Curie temperature. This phase transition is described by an Ising model which formulates the effects of the external magnetic field and interactions between the magnetic spins. However, the uncertainties associated with the ambient temperature and external magnetic field lead to a phase transition zone rather than a single critical phase transition point. To the best of our knowledge, the effects of these external uncertainties, as well as the long-range interactions among the magnetic spins, on the phase transition behavior have never been explicitly modeled. This study presents an Ising model formulation for 2-D magnetic materials by incorporating the effects of long-range interactions between the spins and the external parameter (temperature and external magnetic field) uncertainties on the ferromagnetic-paramagnetic phase transitions.
Multi-scale Crystal Plasticity Model for Superalloys: Shahriyar Keshavarz1; Carelyn Campbell1; Andrew Reid1; 1NIST
This study focuses on modeling creep in superalloys using multi-scale approaches to simulate and predict mechanical responses in crystal plasticity finite element platforms. The model has two important features of morphology including the average size, the volume fraction, and the shape of the precipitates, and composition. The multi-scale framework bridges two sub-grain and homogenized grain scales. For the sub-grain scale, a size-dependent, dislocation density-based constitutive model in the crystal plasticity finite element framework with the explicit depiction of the gamma-gamma prime morphology is used as a building block for the next homogenized scale. For the homogenized scale, a composition-dependent activation energy-based crystal plasticity model is developed which has implicit effects of the morphology. The homogenized model can significantly expedite crystal plasticity FE simulations due to the parameterized representation while retaining accuracy. The glide and climb dislocation mechanisms are proposed in order to capture the creep response of superalloys.