Inference-based Approaches for Material Discovery and Property Optimisation: Structure-Property Inference from Simulations
Sponsored by: TMS Advanced Characterization, Testing, and Simulation Committee, TMS Chemistry and Physics of Materials Committee
Program Organizers: Felix Hofmann, University of Oxford; Michael Short, Massachusetts Institute of Technology; Cody Dennett, Commonwealth Fusion Systems; Mohamed Abdallah Reza, University Of Oxford; Daniel Mason, UK Atomic Energy Authority
Wednesday 2:00 PM
October 12, 2022
Room: 326
Location: David L. Lawrence Convention Center
Session Chair: Michael Short, MIT
2:00 PM Invited
Probing the Local Charge Density and Phonon Dynamics by Electron Microscopy: Xiaoqing Pan1; 1University of California Irvine
The advent of aberration correctors, pixelated direct electron detectors and monochromators marks major milestones in the development of transmission electron microscopy (TEM). The local structure, properties and dynamic behavior of materials can be studied by electron microscopy and spectroscopy. In this talk, a novel 4D STEM diffraction imaging technique is developed to map the local electric field and charge density in real space with sub-Å spatial resolution. With the 4D STEM methods, one can measure the electrical charge density, dipole moment, valence electron distribution in nanostructures and single defects. Furthermore, using the recently developed space- and angle-resolved vibrational electron energy-loss spectroscopy (EELS), we demonstrate a mapping of phonons revealing an interface mode at the Si-Ge interface and phonon dynamics of SiGe quantum dots (QDs). By utilizing averaged and resolved momentum conditions, phonon momenta can be imaged to obtain information about phonon propagation at the nanometer scale.
2:40 PM
A General Solid Solution Strengthening Model in Multicomponent Alloys: Taiwu Yu1; Thomas Barkar2; Paul Mason1; 1Thermo-Calc Software Inc; 2Thermo-Calc Software AB
In the most recent decades, the demand of high-throughput calculations increases significantly for materials designs. An ICME (Integrated Computational Materials Engineering) framework is proposed to predict the solid solution strengthening of multicomponent systems. We adopted Walbrühl’s framework which is based on the concentration dependence of x^(2/3) as proposed by Labusch and fitted the coefficients with experimental and computational simulated results. Furthermore, we calibrated the model and optimized the parameters with over 1000 alloy systems, including a wide range of elements. The calibrated systems include pure metals, binary, ternary and multi-component high entropy alloys. The calibrated model gives a good agreement between the calculated and experimental values. The parameters have been implemented into the Thermo-Calc software to design alloys within the solubility range of the desired phases to obtain properties needed.
3:00 PM
Alloy-agnostic Criteria for Solidification Cracking Susceptibility Evaluation: Rafael Giorjao1; Eric Brizes1; Antonio Ramirez1; 1Ohio State University
Predicting the occurrence of solidification cracking during the solidification of metallic alloys by numerical simulation is a crucial move for avoiding such defects. Several models are widely available, however, the application of such are impacted due to the specific and not accessible parameters required. A simple, composition-based approach to rank solidification cracking susceptibility is presented. The procedure links computational thermodynamic and fluid dynamics to provide an evaluation tool for solidification cracking. The method is related to the liquid filling phenomena in dendritic arms during solidification, which plays a critical role in solidification cracking phenomena. The dendritic profiles were constructed using the fraction of solid calculated by commercial thermodynamic software packages. The method capability to rank the solidification cracking propensity of similar alloys based on composition provides an important new operative tool to aid alloy development in welding and additive manufacturing related areas.
3:20 PM
High Throughput CALPHAD-based Thermodynamic and Kinetic Evaluation of Stainless-steel Solidification: Nathan Daubenmier1; Benjamin Sutton2; Antonio Ramirez1; 1The Ohio State University; 2ThermoCalc
Solidification cracking in stainless steel weld metal poses an additional challenge to fusion-based additive manufacturing processes over traditional manufacturing processes due to the whole part being formed from as solidified material. Because primary solidification mode strongly dictates solidification cracking and is highly dependent on chemical composition and solidification rate, widely used and accepted predictive diagrams such as the Schaeffler and WRC-1992 diagram are utilized to assess potential solidification cracking concerns. Advances in computational techniques enable solidification simulations to be run on large compositional ranges. This provides an additional avenue for predicting primary solidification mode in stainless steel alloys. This work used high-throughput computational thermodynamic simulations to develop a diagram to predict primary solidification mode for stainless steel. By simulating the stable and metastable liquidus temperatures for randomly generated chemistries, a new set of nickel and chromium equivalency relationships were developed that provide sharp delineation between primary austenite and ferrite solidification.
3:40 PM Break
4:00 PM Panel Discussion