An official website of the United States government
Abstract A wide range of deep learning-based machine learning (ML) techniques are extensively applied to the design of high-entropy alloys (HEAs), yielding numerous valuable insights. Kolmogorov–Arnold networks (KAN) is a recently developed architecture that aims to improve both the accuracy and interpretability of input features. In this work, we explore three different datasets for HEA design and demonstrate the application of KAN for both classification and regression models. In the first example, we use a KAN classification model to predict the probability of single-phase formation in high-entropy carbide ceramics based on various properties such as mixing enthalpy and valence electron concentration. In the second example, we employ a KAN regression model to predict the yield strength and ultimate tensile strength of HEAs based on their chemical composition and process conditions including annealing time, cold rolling percentage, and homogenization temperature. The third example involves a KAN classification model to determine whether a certain composition is an HEA or non-HEA, followed by a KAN regressor model to predict the bulk modulus of the identified HEA, aiming to identify HEAs with high bulk modulus. In all three examples, KAN either outperform or match the performance in terms of accuracy such asF1 score for classification and mean square error, and coefficient of determination (R2) for regression of the multilayer perceptron by demonstrating the efficacy of KAN in handling both classification and regression tasks. We provide a promising direction for future research to explore advanced ML techniques, which lead to more accurate predictions and better interpretability of complex materials, ultimately accelerating the discovery and optimization of HEAs with desirable properties.
more »
« less
Disordered enthalpy–entropy descriptor for high-entropy ceramics discovery
Abstract The need for improved functionalities in extreme environments is fuelling interest in high-entropy ceramics1–3. Except for the computational discovery of high-entropy carbides, performed with the entropy-forming-ability descriptor4, most innovation has been slowly driven by experimental means1–3. Hence, advancement in the field needs more theoretical contributions. Here we introduce disordered enthalpy–entropy descriptor (DEED), a descriptor that captures the balance between entropy gains and enthalpy costs, allowing the correct classification of functional synthesizability of multicomponent ceramics, regardless of chemistry and structure. To make our calculations possible, we have developed a convolutional algorithm that drastically reduces computational resources. Moreover, DEED guides the experimental discovery of new single-phase high-entropy carbonitrides and borides. This work, integrated into the AFLOW computational ecosystem, provides an array of potential new candidates, ripe for experimental discoveries.
more »
« less
- PAR ID:
- 10539677
- Author(s) / Creator(s):
- ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; more »
- Publisher / Repository:
- Springer
- Date Published:
- Journal Name:
- Nature
- Volume:
- 625
- Issue:
- 7993
- ISSN:
- 0028-0836
- Page Range / eLocation ID:
- 66 to 73
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Silanes are important in chemistry and material science. The self‐redistribution of HSiCl3is an industrial process to prepare SiH4, which is widely used in electronics and automobile industries. However, selective silane cross‐redistribution to prepare advanced silanes is challenging. We now report an enthalpy‐driven silane cross‐redistribution to access bis‐silanes that contain two different types of Si−H bonds in the same molecule. Compared with entropy‐driven reactions, the enthalpy‐driven reaction shows high regioselectivity, broad substrate scope (62 examples) and high atom economy. Our combined experimental and computational study indicates that the reaction proceeds through a Ni0‐NiII‐NiIVcatalytic cycle.more » « less
-
Abstract Compositionally complex materials have demonstrated extraordinary promise for structural robustness in extreme environments. Of these, the most commonly thought of are high entropy alloys, where chemical complexity grants uncommon combinations of hardness, ductility, and thermal resilience. In contrast to these metal–metal bonded systems, the addition of ionic and covalent bonding has led to the discovery of high entropy ceramics (HECs). These materials also possess outstanding structural, thermal, and chemical robustness but with a far greater variety of functional properties which enable access to continuously controllable magnetic, electronic, and optical phenomena. In this experimentally focused perspective, we outline the potential for HECs in functional applications under extreme environments, where intrinsic stability may provide a new path toward inherently hardened device design. Current works on high entropy carbides, actinide bearing ceramics, and high entropy oxides are reviewed in the areas of radiation, high temperature, and corrosion tolerance where the role of local disorder is shown to create pathways toward self-healing and structural robustness. In this context, new strategies for creating future electronic, magnetic, and optical devices to be operated in harsh environments are outlined.more » « less
-
Herein, we critically evaluate computational and experimental studies in the emerging field of high-entropy ultra-high-temperature ceramics. High-entropy ultra-high-temperature ceramics are candidates for use in extreme environments that include temperatures over 2,000°C, heat fluxes of hundreds of watts per square centimeter, or irradiation from neutrons with energies of several megaelectron volts. Computational studies have been used to predict the ability to synthesize stable high-entropy materials as well as the resulting properties but face challenges such as the number and complexity of unique bonding environments that are possible for these compositionally complex compounds. Experimental studies have synthesized and densified a large number of different high-entropy borides and carbides, but no systematic studies of composition-structure-property relationships have been completed. Overall, this emerging field presents a number of exciting research challenges and numerous opportunities for future studies.more » « less
-
Abstract Piezoelectric materials enable the conversion of mechanical energy into electrical energy and vice‐versa. Ultrahigh piezoelectricity has been only observed in single crystals. Realization of piezoelectric ceramics with longitudinal piezoelectric constant (d33) close to 2000 pC N–1, which combines single crystal‐like high properties and ceramic‐like cost effectiveness, large‐scale manufacturing, and machinability will be a milestone in advancement of piezoelectric ceramic materials. Here, guided by phenomenological models and phase‐field simulations that provide conditions for flattening the energy landscape of polarization, a synergistic design strategy is demonstrated that exploits compositionally driven local structural heterogeneity and microstructural grain orientation/texturing to provide record piezoelectricity in ceramics. This strategy is demonstrated on [001]PC‐textured and Eu3+‐doped Pb(Mg1/3Nb2/3)O3‐PbTiO3(PMN‐PT) ceramics that exhibit the highest piezoelectric coefficient (small‐signald33of up to 1950 pC N–1and large‐signald33* of ≈2100 pm V–1) among all the reported piezoelectric ceramics. Extensive characterization conducted using high‐resolution microscopy and diffraction techniques in conjunction with the computational models reveals the underlying mechanisms governing the piezoelectric performance. Further, the impact of losses on the electromechanical coupling is identified, which plays major role in suppressing the percentage of piezoelectricity enhancement, and the fundamental understanding of loss in this study sheds light on further enhancement of piezoelectricity. These results on cost‐effective and record performance piezoelectric ceramics will launch a new generation of piezoelectric applications.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1038/s41586-023-06786-y
