Algorithm Development in Materials Science and Engineering: Multiscale Algorithms for Crystal Plasticity and Damage Mechanics II
Sponsored by: TMS Materials Processing and Manufacturing Division, TMS Functional Materials Division, TMS Structural Materials Division, TMS: Computational Materials Science and Engineering Committee, TMS: Integrated Computational Materials Engineering Committee, TMS: Phase Transformations Committee, TMS: Solidification Committee, TMS: Chemistry and Physics of Materials Committee
Program Organizers: Adrian Sabau, Oak Ridge National Laboratory; Ebrahim Asadi, University of Memphis; Enrique Martinez Saez, Clemson University; Garritt Tucker, Colorado School of Mines; Hojun Lim, Sandia National Laboratories; Vimal Ramanuj, Oak Ridge National Laboratory
Wednesday 8:30 AM
March 22, 2023
Room: Cobalt 502B
Location: Hilton
Session Chair: Hojun Lim, Sandia National Laboratories; Enrique Saez, Clemson University
8:30 AM Invited
Algorithms for Computing Diffraction Patterns from Dislocation Networks Generated via Discrete Dislocation Dynamics Simulations: Darshan Bamney1; Aaron Tallman2; Laurent Capolungo1; Douglas Spearot3; 1Los Alamos National Laboratory; 2Florida International University; 3University of Florida
X-ray and electron diffraction are common experimental methods used to identify the atomic structure of materials, and can provide approximations of internal defect content, such as dislocation density. The objective of this work is to develop computational algorithms to simulate x-ray diffraction profiles from dislocation networks generated via discrete dislocation dynamics simulations. Two algorithms are developed, both of which leverage the elastic strain fields generated in the presence of dislocations. However, the two algorithms differ in their treatment of correlations in the elastic strain fields; this affects the shape of the computed diffraction peaks. Using these algorithms, a data-driven surrogate modeling method is proposed to correlate diffraction peak broadening to dislocation density. This model is applied to analyze x-ray diffraction peak broadening from dislocation networks in Al and Ta, showing improvement over prior analytical approaches to predict dislocation density.
9:10 AM
An Automated Approach to Data Extraction for SMAs: Dylan Kennedy1; Aaron Stebner1; Branden Kappes2; 1Georgia Institute of Technology; 2KMMD, LLC
Modern materials research has been revolutionized by machine learning (ML) which uses large amounts of data to predict the properties of new materials. The analysis of this data represents a significant bottleneck in the development of ML models. This is especially true for shape memory alloys (SMAs), where phase transformations need to be characterized in addition to standard thermo-mechanical properties. Automating the data extraction process of database building can serve as a solution to this issue, allowing faster model deployment. Additionally, rapidly analyzing experimental data can allow databases to be established from larger and more densely populated search spaces, as these experiments can be easily performed if the desired data does not exist in literature. This work aims to develop a tool that will autonomously extract desired properties of SMAs from raw data collected from a wide array of experiments for use in ML models.
9:30 AM
Development of Structure-property Linkages for Damage in Crystalline Microstructures Using Bayesian Inference and Unsupervised Learning: David Montes De Oca Zapiain1; Anh Tran1; Hojun Lim1; 1Sandia National Laboratories
Crystal Plasticity Finite Element Method (CPFEM) is a robust tool that accurately captures the effects the orientation of the crystal lattice has on the deformation of a material, which can be used to assess damage performance. Nevertheless, the high computational cost of CPFEM makes intractable the usage of this technique in industry. Therefore, there is a critical need for an accurate and computationally-efficient linkage between the internal crystalline texture of a material and its damage performance. This work addresses this critical need by developing an accurate reduced-order model capable of predicting the damage performance of a material from its internal crystalline texture. Furthermore, this work leverages Bayesian inference to optimally select the crystallographic orientations to build an accurate linkage given the fact that each training point requires the evaluation of an expensive high-fidelity CPFEM-simulation. SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.SAND No: SAND2022-8914 A
9:50 AM
Multifaceted Uncertainty Quantification for Structure-property Relationship: Anh Tran1; Pieterjan Robbe1; Hojun Lim1; 1Sandia National Laboratories
Uncertainty quantification (UQ) plays a critical role in verifying and validating forward integrated computational materials engineering (ICME) models. Among numerous ICME models, the crystal plasticity finite element method (CPFEM) is a powerful tool that enables one to assess microstructure-sensitive behaviors and thus, bridge materials structure to performance. Nevertheless, given its nature of constitutive model form and the randomness of microstructures, CPFEM is exposed to both aleatory uncertainty (microstructural variability), as well as epistemic uncertainty (parametric and model-form error). In this work, we discuss several topics in UQ analysis of CPFEM simulations, including forward and inverse UQ problems. In addition, we characterize the effects that various fidelity parameters, such as mesh resolutions, integration time-steps, and constitutive models have in CPFEM with our UQ analysis.
10:10 AM Break
10:25 AM
Novel Multi-scale Plasticity Modeling Using Defect Dynamics Element Method (DDEM): Nicole Aragon1; Dongchan Jang2; Hojun Lim1; Ill Ryu3; 1Sandia National Laboratories; 2Korea Advanced Institute of Science and Technology; 3The University of Texas at Dallas
A multi-scale modeling technique has been developed by coupling dislocation dynamics (DD) and finite element modeling, known as the defect dynamics element method (DDEM). In this coupled model, the DD framework tracks the evolution of the dislocation microstructure, while the stress field with the given plastic strain field from DD is calculated in the finite element model. In this talk, applications of DDEM will be presented with the aim of correlating the defect microstructure characteristics and the macroscopic material behavior under quasi-static and dynamic loading conditions. Micro-bending simulation results of single crystalline and bicrystalline copper microbeams show good agreement with performed experiments. For the Taylor impact test with significant temperature and strain rate gradients, DDEM simulations of single crystalline tantalum predict a similar anisotropic response to that observed in previous experimental findings. This framework enables robust multi-scale plasticity modeling for complex loading and boundary conditions as well as multi-physical phenomena.
10:45 AM
Multiphase Microstructure-based Modeling for Rolling Contact Fatigue Life Prediction: Jinheung Park1; Kijung Lee1; Soonwoo Kwon2; Myoung-Gyu Lee1; 1Seoul National University; 2Hyundai Motor Company
Bearing steels typically consist of martensite and retained austenite to withstand the rolling contact fatigue (RCF) environments. Experiments report that the subsurface microstructure of bearing steels often exhibits deformation-induced martensitic transformation of retained austenite and mechanical softening caused by microstructural alterations under RCF process. These microstructural characteristics directly affect the fatigue resistance of bearing steel and can increase the uncertainty of fatigue behavior. In this study, a RCF life prediction model is developed considering the experimentally observed microstructural characteristics in martensitic steel. Virtual multiphase microstructure with hierarchically structured martensite and austenite phases is developed. Then, the deformation-induced martensitic transformation and dislocation-assisted carbon migration are numerically implemented in crystal plasticity (CP) finite element framework. The CP model is then coupled with continuum damage mechanics to predict crack initiation, propagation, and failure during RCF. Finally, the developed model is validated by predicting the Weibull distribution for RCF life with a probabilistic approach.
11:05 AM
Prediction of Mechanical Properties in a Bulged and Annealed Steel Tube through a Multiscale Modeling Approach Based on CPFEM: Amir Asgharzadeh1; Taejoon Park1; Farhang Pourboghrat1; 1The Ohio State University
A multiscale modeling approach based on CPFEM is proposed to predict the mechanical properties in the hydroformed and subsequently annealed steel tube. To that end, CPFEM modeling of the deformation behavior in metals consisting of second phase particles is performed based on RVE models. The microstructure based RVE model for the as-received tubular material is generated based on the experimental data obtained from microstructural observations, including the configuration of precipitates, grain topology and orientation distribution. Using the generated RVE model, the CP model is calibrated against the experimental tensile data and then the deformation behavior during THF is predicted. The microstructural evolution during subsequent annealing process is predicted through Cellular Automata model and incorporated into the RVE model to represent the material at different annealed conditions. The virtual tensile test is performed on the RVE models corresponding to different stages of the process to assess the evolution of mechanical properties.
11:25 AM
Symmetry Relation Database and Its Application to Ferroelectric Materials Discovery: Qiang Zhu1; Byungkyun Kang1; Kevin Parrish1; 1University of Nevada, Las Vegas
The ability to understand the atomistic mechanisms that occur in the solid phase transition is of crucial importance in materials research. To investigate the displacive phase transition at the atomic scale, we have implemented a numerical algorithm to automate the detection the symmetry relations between any two candidate crystal structures. Using this algorithm, we systematically screen all possible polar-nonpolar structure pairs from the entire Materials Project database and establish a database of ~4500 pairs that possess a close symmetry relation. These pairs can be connected through a continuous phase transition with small atomic displacements. From this database, we identify several new ferroelectric materials that have never been reported in the past. In addition to the screening of ferroelectric materials, the symmetry relation database may also be used for other areas, such as material structure prediction and new materials discovery and X-ray diffraction analysis.
11:45 AM
Coupling of a Multi-GPU Accelerated Elasto-visco-plastic Fast Fourier Transform Constitutive Model with the Implicit Finite Element Method: Marko Knezevic1; 1University of New Hampshire
This paper presents an implementation of the elasto-visco-plastic fast Fourier transform (EVPFFT) crystal plasticity model in the implicit finite element (FE) method through a user material (UMAT) subroutine of Abaqus standard. The constitutive response at every integration point is obtained by the full-field homogenization over an explicit microstructural cell. The implementation is a parallel computing approach involving multi-core central processing units (CPUs) and graphics processing units (GPUs) for computationally efficient simulations of large plastic deformation of metallic components with arbitrary geometry and loading boundary conditions. To this end, the EVPFFT solver takes advantages of GPU acceleration utilizing Nvidia’s high performance computing software development kit (SDK) compiler and compute unified device architecture (CUDA) FFT libraries, while the FE solver leverages the message passing interface (MPI) for parallelism across CPUs. Several benchmark and application case studies will be presented to illustrate the potential and utility of the developed multi-level simulation strategy.