HEA 2023: Fundamental Theory and Modeling III
Program Organizers: Andrew Detor, DARPA/DSO; Amy Clarke, Los Alamos National Laboratory
Tuesday 1:40 PM
November 14, 2023
Room: Three Rivers
Location: Omni William Penn
Session Chair: Guofeng Wang, University of Pittsburgh
1:40 PM Introductory Comments
1:45 PM Invited
Machine Learning Guided High Entropy Alloy Development: John Sharon1; Ken Smith1; Ryan Deacon1; Anthony Ventura1; Soumalya Sarkar1; GV Srinivasan1; 1Raytheon Technologies Research Center
High Entropy Alloys (HEAs) with multiple principal elements have demonstrated enhanced properties that can rival or exceed conventional alloy systems. HEAs are typically comprised of 4 or more elements present from 5 to 35 at.% resulting in a large combinatorial composition space for which computational tools are vital to sort through combinations and identify the most promising candidates. A variety of analytical and other relatively fast computational models are available to help identify candidates. This talk will describe a machine learning framework assembled to assist in identifying candidates that leverages experimental data, published literature, as well as mechanistic models. Examples of using the framework to identify potential HEA candidates will be provided along with complementary experimental characterization and validation. Industry perspective on HEA maturation and adoption for engineering applications will also be discussed
2:15 PM
Grain Boundary Segregation in High Entropy Alloys: Theoretical Development and Mesoscale Simulations: Fadi Abdeljawad1; Milad Taghizadeh1; Malek Alkayyali1; 1Clemson University
Owing to their far-from-dilute compositions, high entropy alloys (HEAs) exhibit unique properties that are not typically encountered in conventional alloys. Recent experiments revealed segregation of elemental species to grain boundaries (GBs) in several HEAs. Even minute amounts of alloys at GBs greatly influence boundary dynamics, including GB migration and grain coarsening. Herein, we present a theoretical and computational model of GB segregation in HEAs and its impact on GB migration. The model accounts for bulk thermodynamics and the interaction of various elemental species with GBs, and it captures various mass transport processes. Simulation results reveal a plethora of segregation behaviors, including synergistic co-segregation and induced de-segregation, that are dependent on alloy-alloy interactions within the GB. Our approach provides avenues to employ GB segregation engineering as a strategy to design HEAs with tailored microstructures.
2:35 PM
Developing CALPHAD Databases for High Entropy Alloys: Huahai Mao1; Martin Xing1; Reza Naraghi1; Qing Chen1; Paul Mason2; 1Thermo-Calc Software AB; 2Thermo-Calc Software Inc.
CALPHAD modeling of High Entropy Alloys presents unique challenges compared with that of conventional alloys due to the lack of a single principal element. It requires an accurate description for the entire composition range rather than just in the vicinity of the base element. Here we describe the approach taken in developing the TCHEA thermodynamic database, which contains 26 elements, where almost all underlying binary and over 500 ternary systems have been critically evaluated to capture the composition and temperature dependence. We also present the corresponding MOBHEA mobility database. The CALPHAD approach to describing composition and temperature dependence can also be extended to many other thermochemical or thermophysical properties such as density, viscosity, surface tension, thermal conductivity, and electrical resistivity. All these properties are modelled in the TCHEA database. Validation with experimental data for various HEAs and application examples will also be given.
2:55 PM
Phase-field Simulation of Phase Separation in MPEAs: Kamalnath Kadirvel1; Weisheng Cao1; Shuanglin Chen1; Yunzhi Wang2; Shalini Koneru2; Fan Zhang1; 1Computherm LLC; 2Ohio State University
MPEAs (Multi Principal Element Alloys) owing to the compositional complexity can have convoluted phase transformation pathways (PTPs). Understanding these PTPs is critical for tailoring the microstructure for specific engineering applications. We utilized the recently developed phase-field simulation tool (called PanPhasefield, https://computherm.com/panphasefield ) that efficiently couples with CALPHAD databases for thermodynamic and kinetic data to simulate the spinodal decomposition in MPEAs. We studied kinetics of phase separation in Fe-Co-Mn-Ni-Cu alloy system using PanHEA database. Interestingly, the partitioning of Ni and Mn is much slower compared to other elements in the early stages of decomposition despite their relatively high atomic mobilities. However, at the later stages of decomposition, the concentration modulations of Ni and Mn were comparable to those of other elements such as Fe and Co. Simulated results were in qualitative agreement with experiments. Independent thermodynamic calculations were also performed to decouple the effect of free energy and atomic mobility.
3:15 PM Break
3:35 PM
A Computational Thermodynamics Framework with Intrinsic Short-range Order: Chu-Liang Fu1; Bi-Cheng Zhou1; 1University of Virginia
CALPHAD is a leading method for modeling and calculations of phase equilibria in materials. However, the prevailing solution model used in CALPHAD, the sublattice model, is an empirical mean-field model based on the ideal entropy approximation, which makes CALPHAD inadequate for describing chemical short-range order (SRO) in alloys. Here we develop a hybrid framework by marrying advantages from the Cluster Variation Method and CALPHAD through incorporating chemical SRO into CALPHAD with a novel cluster-based solution model. We have put more physics into CALPHAD, while maintaining its practicality and efficiency. The configurational and non-configurational (vibrational, elastic, electronic) contributions to free energy are modeled separately, gaining insights into their respective effects on phase stability. Phase diagrams of representative alloy systems are calculated, showing great comparison with experiments. This hybrid CVM-CALPHAD framework represents a new methodology for thermodynamic modeling that enables SRO to be exploited for the design of novel complex concentrated alloys.
3:55 PM
Prediction of Atomic Distribution in Solid Solutions via CALPHAD-based Models: Shalini Roy Koneru1; Kamal Kadirvel2; Hamish Fraser1; Yunzhi Wang1; 1Ohio State University; 2Computherm LLC
With the advent of HEAs, there is an increased interest over the prediction of solid solution phase properties. The atomic distribution in solid solutions, i.e., if it is truly random or not, can affect the mechanical, electrical and magnetic properties. Thus, researchers attempt to predict the atomic distributions in solid solutions through calculation of interatomic interaction energies via atomistic simulations. However, the interatomic interaction energies are also inherently present in the CALPHAD databases. Thus, we developed a CALPHAD module to predict the atomic distributions by directly utilizing the existing commercial databases. Here, we extended the de Fontaine’s theory and utilized the inverse of the Hessian of the free energy of the solid solution (in the single solid solution region above any miscibility gap) to calculate the pair correlation functions and thereby short-range order (SRO) parameters. We will demonstrate the model via SRO prediction in different binary and ternary alloys.
4:15 PM
Design of High Hardness Carbide Reinforced High-entropy Alloys: Joshua Berry1; Yunus Azakli1; Olivier Messe2; Katerina Christofidou1; Iain Todd1; 1University Of Sheffield; 2Oerlikon AM Europe GmbH
High Entropy Alloys (HEAs) present an opportunity for the design and development of new wear resistant hardmetals, to replace the conventional WC-Co cemented carbides, used in demanding metal forming applications. Suitable alloy systems require high hardness and wear resistance, including at high temperatures, while retaining ductility for damage tolerance. To streamline the search for a new high-entropy alloy system capable of satisfying the design constraints, a combined machine learning and CALPHAD approach has been undertaken, targeting FCC systems. A set of nine of the hardest predicted FCC solid solution forming HEA compositions from the machine learning model were selected and fabricated, from which four demonstrated suitable microstructure for further carbon reinforcement. Mechanical and thermal assessment of these carbide reinforced alloys will be presented and rationalised through CALPHAD.This work was supported by Oerlikon AM Europe GmbH, Engineering and Physical Sciences Research Council UK [EP/S022635/1] and Science Foundation Ireland [18/EPSRC-CDT/3584].
4:35 PM
Rapid Design of Eutectic and Ordered HEAs using The Alloy Optimization Software (TAOS): Nicholas Ury1; Aurelien Perron1; Brandon Bocklund1; Thomas Voisin1; Vincenzo Lordi1; Joseph Mckeown1; 1LLNL
High entropy alloys (HEAs), or multi-principal complex alloys (MPEAs) unlock the composition space from traditionally single-element rich alloys and offer a vast landscape of undiscovered alloys. However, when applying constraints during the alloy design process, the actual composition space becomes much smaller. Furthermore, finding this subset of compositions as well as optimizing an alloy for certain properties by hand becomes tedious. The Alloy Optimization Software (TAOS) was designed to overcome this issue by leveraging Calphad-based software packages such as Thermo-Calc and pycalphad coupled to blackbox optimizers. TAOS allows many types of objectives and constraints to be used for optimization. Two case studies focusing on eutectic and ordered HEAs are shown and the design process, lessons learned, and experimental results of the new alloy compositions will be discussed.
4:55 PM
Thermodynamics of Refractory Compositionally Complex Alloys: Eric Lass1; 1University of Tennessee-Knoxville
From high temperature materials that may replace Ni-based superalloys to lightweight structural alloys, refractory compositionally complex alloys (R-CCAs) containing elements from Groups IV through VI of the periodic table represent an exciting new class of alloys for future exploitation in next generation technologies. R-CCAs are most often single-phase BCC, or a majority BCC also containing secondary phases such as B2, Laves, and others. BCC alloys behave fundamentally different than FCC-based alloys, such as Al- or Ni-based superalloys, including the thermodynamically higher-order BCC-B2 phase relationship and thermally activated dislocation motion governing plastic deformation even at room temperature, which leads to the ductile-to-brittle transition exhibited by most BCC materials. This work explores the thermodynamics of these materials, including the implications of the higher-order nature of the BCC-B2 transformation on precipitation strengthening and the effects of thermodynamics on plastic deformation and transformation-induced plasticity (TRIP) behavior.