High Entropy Alloys VIII: Modeling and Machine Learning
Sponsored by: TMS Functional Materials Division, TMS Structural Materials Division, TMS: Alloy Phases Committee, TMS: Mechanical Behavior of Materials Committee
Program Organizers: Peter Liaw, University of Tennessee; Michael Gao, National Energy Technology Laboratory; E-Wen Huang, National Chiao Tung University; Srivatsan Tirumalai; Xie Xie, FCA US LLC; Gongyao Wang, Globus Medical
Thursday 8:30 AM
February 27, 2020
Room: Marina Ballroom E
Location: Marriott Marquis Hotel
Session Chair: Wei Chen, University At Buffalo; Baldur Steingrimsson, Imagars LLC
8:30 AM Invited
Atomistic Simulations of Dislocations and Deformation in FCC Ni-based Binary Concentrated Alloys: Haixuan Xu1; Liubin Xu1; Jaswanth Bommidi1; 1University of Tennessee Knoxville
The emergence of concentrated alloys, such as medium/high entropy alloys, has attracted tremendous research interest, and also posted challenges to traditional solid-solution strengthening theory. To extend the theory, we carried out atomistic simulations to determine the Peierls stress across a wide compositional range in the selected Ni-based binary concentrated alloys. The Peierls stress is found to be influenced by local chemical environments, and the compositional dependence deviates from either strong or weak interaction models. Moreover, we performed nanoindentation simulations to understand the effects of alloy compositions on deformation behavior. The dislocation evolution during the atomistic nanoindentation simulations is examined. The proportion of different types of dislocations generated is associated with the stacking fault energies (SFEs) of the alloys, which may provide insights in understanding the strengthening of these alloys.
8:50 AM Invited
Machine-learning Driven Efficient Exploration of the High Entropy Alloy Phase Space: Raymundo Arroyave1; 1Texas A&M University
High Entropy Alloys are alloys that contain multiple principal alloying elements. While many HEAs have been shown to have unique properties, their discovery has been largely done through costly and time-consuming trial-and-error approaches, with only an infinitesimally small fraction of the entire possible composition space having been explored. In this talk, we will present a recently developed framework that has mapped the problem of exploring the HEA space to a Constraint Satisfaction Problem (CSP), whose solution is, in turn, a one-class classifier implemented as a Support Vector Data Descriptor (SVDD). The resulting algorithm is used to discover regions in the HEA Composition-Temperature space that satisfy desired phase constitution requirements that essentially are mathematical representations of alloy specifications. The framework is capable of identifying regions in the HEA space with arbitrary phase constitution attributes. We demonstrate, as an example, the targeted discovery of precipitation strengthened HEAs.
9:10 AM Invited
Characteristics of Edge Dislocations and their Glide in FCC NiCoFe and NiCoFeCu Equiatomic Solid Solution Alloys: Wei Li1; Satish Rao1; Jaafar El-Awady1; 1Johns Hopkins University
The critical stresses and core structures of 1/2<110> edge dislocations in face centered cubic NiCoFe and NiCoFeCu equiatomic solid solution alloys is investigated using both molecular statics and molecular dynamics simulations at different temeratures (10, 150 and 300K). The computed intrinsic stacking fault energies (ISF) was predicted to be 17 and 28 mJ/m^2 for the NiCoFe and NiCoFeCu alloy, respectively. The dislocation stacking fault width in both crystals is shown to vary along the dislocation length due to fluctuations in the local atomic concentrations. The critical stress for dislocation glide as a function of temperature was also predicted for both alloys and the characteristics of dislocation glide will be discussed. Finally, the results are discussed in terms of a theoretical model for solid solution strengthening.
9:30 AM
Origins of Excellent Passivation in Multiple Principle Element Alloys: John Scully1; Angela Gerard1; Carol Glover1; Kang Wang1; Bi-Cheng Zhou1; Prasanna Balachandran1; Gerald Frankel2; Pin Lu3; James Saal3; Daniel Schreiber4; Joseph Poon1; Sean Agnew1; 1University of Virginia; 2Ohio State University; 3QuesTek Innovations LLC; 4Pacific Northwest National Laboratory
Multiple principle element alloys (MPEAs) are emerging materials that have recently attracted great interest due to their many degrees of freedom in alloy design and synthesis leading to potentially attractive properties. Many unique or even heretofore unobtainable properties have been reported including excellent resistance to various forms of degradation in harsh environments. These include promising aqueous corrosion behavior, improved oxidation resistance in high and intermediate temperature atmospheres, irradiation resistance as well as possible resistance to environmental assisted cracking. Focusing on passivation regulated by the formation and growth of a thin oxide film, a rich variety of possible protective oxide characteristics and attributes are discussed. Both thermodynamic and kinetic factors are considered to develop a better understanding of the key aspects governing the protective nature of these oxide films.
9:50 AM Invited
Connecting Chemical and Structural Order in High Entropy Alloys: Daniel Foley1; James Hart1; Elaf Anber2; Robert Ritchie3; Andrew Minor3; Mark Asta3; Flynn Walsh3; Michael Titus4; Jean-Philippe Couzinie5; Mitra Taheri1; 1Johns Hopkins University; 2Drexel University; 3University of California, Berkeley/Lawrence Berkeley National Laboratory; 4Purdue University; 5University Paris-Est Créteil (UPEC)
This talk reviews recent work complex systems local structure evolution in high entropy alloys (HEAs). Despite their nominal chemical disorder, several studies have reported short range order (SRO) in HEAs – i.e. preferential bonding, local elemental enrichment and/or clustering – and such SRO may have broad implications for HEA performance. To tackle this problem, a suite of spatially resolved, electron imaging, diffraction, and spectroscopy techniques is leveraged to correlate local order with microstructural evolution and related dislocation phenomena. Microstructures of HEAs subjected to a variety of deformation regimes and quantified using diffraction-based techniques. These are compared to simulations of similar alloy families in order to determine the extent to which localized dislocation-based phenomena play a role in microstructural evolution, and how alloy chemistry plays a role in these determining factors. The techniques presented allow for the direct observation of the interplay between chemistry and microstructure, and thus, provides us with key tuning knobs for future HEA development.
10:10 AM Break
10:30 AM
Unravelling Sluggish Diffusion of High-entropy Alloys through Machine Learning Methods: S. Mohadeseh Taheri-Mousavi1; S. Sina Moeini-Ardakani1; Ryan Penny1; Ju Li1; A. John Hart1; 1Massachusetts Institute of Technology
High-entropy alloys (HEAs) can exhibit an exceptional combination of superior damage tolerance and strength at extreme temperatures. Since the advent of HEAs, their single-phase stabilization and the attributed mechanical properties are postulated to be enabled by sluggish diffusion of the alloying elements. Yet, due to the compositional complexity of HEAs, this hypothesis has never been systematically investigated and confirmed. Here, we present a newly-developed numerical framework whereby a machine-learning algorithm supervised by atomistic-scale simulations is used to explore the nanoscale features controlling the diffusivity of alloying components in HEAs. Analysis of all possible atomic configurations within a model HEA by the trained algorithm reveals how the size, and cohesive energy of alloying elements alter the diffusivity rate of the material. In the future, this understanding can be used to guide conventional processing or additive manufacturing, and could enable design of metals with tailored gradient diffusivity.
10:50 AM
Microstructure and Mechanical Properties of Multi-phase CrMnFeCoNiAl0.75 High Entropy Alloy at Intermediate Temperature: Lijing Lin1; Zhihong Zhong1; Peter Liaw2; 1Hefei University of Technology; 2University of Tennessee, Knoxville
A multi-phase high-entropy alloy CrMnFeCoNiAl0.75 was found to be composed of the face-centered-cubic (FCC) ,body-centered-cubic (BCC) and cuboidal B2 structures, and exhibited the high strength and acceptable ductility both at room and elevated temperatures for structural applications. The alloy had a high compression yield strength of 1.194 GPa and limited plasticity of 2.72% at room temperature, while it showed a high yield strength and adequate plasticity at elevated temperatures up to 500 ℃, the value of which was 1.070 GPa and 15.30 %, respectively. The excellent mechanical properties of the CrMnFeCoNiAl0.75 alloy were attributed to a superalloy-like microstructure, which was characterized by cuboidal B2 precipitates coherently embedded in the BCC channel. Further increase in temperature resulted in pronounced softening.
11:10 AM Invited
Materials Fingerprint Classification: Vasileios Maroulas1; Adam Spannaus1; David Keffer1; Kody Law2; Farzana Nasrin1; Cassie Micucci1; Peter Liaw1; Piotr Luszczek1; Louis Santodonato3; 1University of Tennessee; 2University of Manchester; 3Advanced Research Systems, Inc.
State-of-the-art methods for visualizing the local atomic structure, such as atom probe tomography (APT), create noisy and sparse datasets comprised of millions of atoms, but are currently unable to determine the lattice structure. Viewed through the lens of topological data analysis, the essential differences between body-centered and face-centered crystal structures are revealed, allowing researchers to discern materials composed of either lattice type from the APT data. In this talk, we describe how to characterize these fundamental building blocks via topological descriptors and how we combine these descriptors to create an accurate prediction algorithm. Using a new paradigm in computationally-driven materials science, we propose a materials fingerprint -- a machine learning methodology for determining the crystal structure of a material from a noisy and sparse dataset. Using this fingerprinting method, the crystal structures of high-entropy alloys are classified with near perfect accuracy using the APT data.
11:30 AM Invited
Machine Learning for Accelerating the Design of Additively-manufactured Turbine Blades Yielding Ultra-high Energy Efficiency: Xuesong Fan1; Baldur Steingrimsson2; Duckbong Kim3; Peter Liaw1; 1University of Tennessee; 2Imagars LLC; Portland State University; 3Tennessee Tech University
This abstract addresses the application of machine learning (ML) to accelerate design of high-entropy alloys (HEAs), exhibiting exceptional material properties, especially at high temperatures. We address development of useful inverse design representations, and advanced physics-based metallurgical models, enabling identification of HEAs suitable for compressor blades of land-based turbines operating with ultra-high efficiency. By operating turbines with blades made of refractory HEAs at significantly-higher temperatures than conventional alloys, one can expect drastic improvements in efficiency. Assuming fossil-fuel combustion, the first stage of a modern turbine (the stage directly following the combustor) typically faces temperatures around 2,500°F (1,370°C). Modern military jet engines, like the Snecma M88, can experience turbine temperatures of 2,900°F (1,590°C). These high temperatures weaken the blades and make them more susceptible to creep failures. The high temperatures can also make the blades susceptible to corrosion failures. Finally, vibrations from the engine and the turbine itself can cause fatigue failures.
11:50 AM Invited
Quantifying Short-range Ordering in Refractory High-entropy Alloys: Wei Chen1; George Kim1; Chanho Lee2; Peter Liaw2; 1Illinois Institute of Technology; 2University of Tennessee
The material-design strategy of combining multiple elements in near-equimolar ratios has spearheaded the emergence of high-entropy alloys (HEAs), an exciting class of materials with exceptional engineering properties. While random mixing has been widely assumed in multi-principal element solid solutions, both experimental and computational evidence suggests short-range ordering (SRO) exists in many solid-solution HEAs. We employed an integrated first-principles and experimental approach to understand the thermodynamic effects of SRO in the refractory NbTaTiV and NbTaTiVZr HEA systems. Results are compared with predictions from Special Quasi-random Structures (SQS). The existence of SRO produces distinct lattice distortion features in these HEAs and affects their mechanical properties.