Design high-entropy carbide ceramics from machine learning

Abstract

High-entropy ceramics (HECs) have shown great application potential under demanding conditions, such as high stresses and temperatures. However, the immense phase space poses great challenges for the rational design of new high-performance HECs. In this work, we develop machine-learning (ML) models to discover high-entropy ceramic carbides (HECCs). Built upon attributes of HECCs and their constituent precursors, our ML models demonstrate a high prediction accuracy (0.982). Using the well-trained ML models, we evaluate the single-phase probability of 90 HECCs that are not experimentally reported so far. Several of these predictions are validated by our experiments. We further establish the phase diagrams for non-equiatomic HECCs spanning the whole composition space by which the single-phase regime can be easily identified. Our ML models can predict both equiatomic and non-equiatomic HECs based solely on the chemical descriptors of constituent transition-metal-carbide precursors, which paves the way for the high-throughput design of HECCs with superior properties.

High‐Entropy Catalyst—A Novel Platform for Electrochemical Water Splitting

“…Interestingly, ML models are available to forecast the entropy‐forming ability for rationally designing high‐entropy ceramic carbides based on DFT, and two well‐trained models − artificial neural network (ANN) and support vector machine (SVM) − have been used satisfactorily to screen the feasible elemental combinations. [ 107 ]…”

Stabilization, Characterization, and Electrochemical Applications of High‐Entropy Oxides: Critical Assessment of Crystal Phase–Properties Relationship

“…The ML method achieves simulations with comparable accuracy and exhibits better flexibility in the dynamic nature of phase formation and stability than the conventional DFT method. [39] This model has been developed to explore and predict material composition, phase formation, mechanical property, and application performance in the past few years. [40] Zhang et al reported ML models to evaluate the single-phase probability of 90 possible high entropy carbide ceramics (HECCs) composed of group IV-VI metals.…”

“…[40] Zhang et al reported ML models to evaluate the single-phase probability of 90 possible high entropy carbide ceramics (HECCs) composed of group IV-VI metals. [39] The designed models combined with the preliminary DFT results screened 38 single-phase HECCs with high prediction accuracy (0.982). Furthermore, the group revealed that the valence electron concentration (VEC) deviations and the electronegativity deviations of the constituents were critical factors for the phase formation of HECC.…”

Ni-free high-entropy rock salt oxides with Li superionic conductivity

“…This finding opened a kind of “wild west” for ceramists currently racing to discover new materials taking advantage of both experimental and analytical approaches. 9,33,34…”

“…This nding opened a kind of "wild west" for ceramists currently racing to discover new materials taking advantage of both experimental and analytical approaches. 9,33,34 Of special interest are high-entropy oxides (HEO) with rock salt structures. In particular, alkali-doped rock salts in the MgO-CoO-NiO-CuO-ZnO system have shown surprising electrochemical and dielectric properties, including giant dielectric constant, 35 catalytic 36 and electrocatalytic activity, 37 Lisuperionic conductivity 38 and high Li-storage capacity.…”