Abstract:We propose a neural evolution structure (NES) generation methodology combining artificial neural networks and evolutionary algorithms to generate high entropy alloy structures. Our inverse design approach is based on pair distribution functions and atomic properties and allows one to train a model on smaller unit cells and then generate a larger cell. With a speed-up factor of ∼1000 with respect to the special quasi-random structures (SQSs), the NESs dramatically reduce computational costs and time, making pos… Show more
“…Neural evolution structure (NES) generator combines evolutionary algorithms with artificial neural networks (ANNs) to generate MPEA structures. 21 Figure 1 illustrates the workflow for NES structure generation outlining the NES training process and generation process. Here an ANN is trained using a smaller system, and a larger system is generated using this trained model.…”
Section: Atomistic Calculationsmentioning
confidence: 99%
“… Figure 1 A sketch of the required algorithms for the NES training process (A) NES training process (B) NES generation process (C) Outline of the mutation step. Reprinted from 21 with the permission of AIP publishing. …”
Section: Atomistic Calculationsmentioning
confidence: 99%
“… (A) NES training process (B) NES generation process (C) Outline of the mutation step. Reprinted from 21 with the permission of AIP publishing. …”
Section: Atomistic Calculationsmentioning
confidence: 99%
“…This method can be used to create a system with hundreds of thousands of atoms. 22 A combination of NES 21 and OTIS 22 methods can be used to create MPEA structures that could be used in large-scale atomistic simulations.…”
Section: Atomistic Calculationsmentioning
confidence: 99%
“…Recent studies discussed in a previous section have proposed novel methods to tackle this problem and generate reliable structures with a high level of confidence. 21 , 22 Still, there is a significant disconnect within the structure generation algorithms while considering the thermodynamic stability of the structures that need to be addressed.…”
“…Neural evolution structure (NES) generator combines evolutionary algorithms with artificial neural networks (ANNs) to generate MPEA structures. 21 Figure 1 illustrates the workflow for NES structure generation outlining the NES training process and generation process. Here an ANN is trained using a smaller system, and a larger system is generated using this trained model.…”
Section: Atomistic Calculationsmentioning
confidence: 99%
“… Figure 1 A sketch of the required algorithms for the NES training process (A) NES training process (B) NES generation process (C) Outline of the mutation step. Reprinted from 21 with the permission of AIP publishing. …”
Section: Atomistic Calculationsmentioning
confidence: 99%
“… (A) NES training process (B) NES generation process (C) Outline of the mutation step. Reprinted from 21 with the permission of AIP publishing. …”
Section: Atomistic Calculationsmentioning
confidence: 99%
“…This method can be used to create a system with hundreds of thousands of atoms. 22 A combination of NES 21 and OTIS 22 methods can be used to create MPEA structures that could be used in large-scale atomistic simulations.…”
Section: Atomistic Calculationsmentioning
confidence: 99%
“…Recent studies discussed in a previous section have proposed novel methods to tackle this problem and generate reliable structures with a high level of confidence. 21 , 22 Still, there is a significant disconnect within the structure generation algorithms while considering the thermodynamic stability of the structures that need to be addressed.…”
At present, alloys have broad application prospects in heterogeneous catalysis, due to their various catalytic active sites produced by their vast element combinations and complex geometric structures. However, it is the diverse variables of alloys that lead to the difficulty in understanding the structure‐property relationship for conventional experimental and theoretical methods. Fortunately, machine learning methods are helpful to address the issue. Machine learning can not only deal with a large number of data rapidly, but also help establish the physical picture of reactions in multidimensional heterogeneous catalysis. The key challenge in machine learning is the exploration of suitable general descriptors to accurately describe various types of alloy catalysts, which help reasonably design catalysts and efficiently screen candidates. In this review, several kinds of machine learning methods commonly used in the design of alloy catalysts is introduced, and the applications of various reactivity descriptors corresponding to different alloy systems is summarized. Importantly, this work clarifies the existing understanding of physical picture of heterogeneous catalysis, and emphasize the significance of rational selection of universal descriptors. Finally, the development of heterogeneous catalytic descriptors for machine learning are presented.
To achieve an equitable energy transition toward net-zero 2050 goals, the electrochemical reduction of CO 2 (CO 2 RR) to chemical feedstocks through utilizing both CO 2 and renewable energy is particularly attractive. However, the catalytic activity of CO 2 RR is limited by the scaling relation of the adsorption energies of intermediates. Circumventing the scaling relation is a potential strategy to achieve a breakthrough in catalytic activity. Herein, based on density functional theory (DFT) calculations, we designed a high-entropy alloy (HEA) system of FeCoNiCuMo with high catalytic activity for CO 2 RR. Machine learning models were developed by considering 1280 adsorption sites to predict the adsorption energies of COOH*, CO*, and CHO*. The scaling relation between the adsorption energies of COOH*, CO*, and CHO* is circumvented by the rotation of COOH* and CHO* on the designed HEA surface, resulting in the outstanding catalytic activity of CO 2 RR with the limiting potential of 0.29−0.51 V. This work not only accelerates the development of HEA catalysts but also provides an effective strategy to circumvent the scaling relation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
