scholarly journals Atomic Structure-free Representation of Active Motifs for Expedited Catalyst Discovery

2021 ◽  
Author(s):  
Dong Hyeon Mok ◽  
Seoin Back
Keyword(s):  
Atomic Structure ◽  
Density Functional ◽  
Nearest Neighbor ◽  
Catalyst Surface ◽  
Binding Energies ◽  
Reasonable Accuracy ◽  
Density Functional Calculation ◽  
Energy Prediction ◽  
Binding Energy Prediction ◽  
Free Representation

For CO* and H* binding energy prediction, we develop new representation of catalyst surface which split surface into three types of site, first nearest neighbor of adsorbates and second nearest neighbor in same layer and sublayer. From this representation and machine learning regression model, we achieve reasonable accuracy (0.120 eV for CO* and 0.105 eV for H*) with quick training (~200 sec using CPU). Because our representation does not require density functional calculation and atomic structure modelling, it can predict binding energies of possible active motifs without time-consuming steps.

scholarly journals Atomic Structure-free Representation of Active Motifs for Expedited Catalyst Discovery

2021 ◽  
Author(s):  
Dong Hyeon Mok ◽  
Seoin Back
Keyword(s):  
Atomic Structure ◽  
Density Functional ◽  
Nearest Neighbor ◽  
Catalyst Surface ◽  
Binding Energies ◽  
Reasonable Accuracy ◽  
Density Functional Calculation ◽  
Energy Prediction ◽  
Binding Energy Prediction ◽  
Free Representation

For CO* and H* binding energy prediction, we develop new representation of catalyst surface which split surface into three types of site, first nearest neighbor of adsorbates and second nearest neighbor in same layer and sublayer. From this representation and machine learning regression model, we achieve reasonable accuracy (0.120 eV for CO* and 0.105 eV for H*) with quick training (~200 sec using CPU). Because our representation does not require density functional calculation and atomic structure modelling, it can predict binding energies of possible active motifs without time-consuming steps.

scholarly journals Atomic Structure-free Representation of Active Motifs for Expedited Catalyst Discovery

2021 ◽  
Author(s):  
Dong Hyeon Mok ◽  
Seoin Back
Keyword(s):  
Atomic Structure ◽  
Density Functional ◽  
Nearest Neighbor ◽  
Catalyst Surface ◽  
Binding Energies ◽  
Reasonable Accuracy ◽  
Density Functional Calculation ◽  
Energy Prediction ◽  
Binding Energy Prediction ◽  
Free Representation

For CO* and H* binding energy prediction, we develop new representation of catalyst surface which split surface into three types of site, first nearest neighbor of adsorbates and second nearest neighbor in same layer and sublayer. From this representation and machine learning regression model, we achieve reasonable accuracy (0.120 eV for CO* and 0.105 eV for H*) with quick training (~200 sec using CPU). Because our representation does not require density functional calculation and atomic structure modelling, it can predict binding energies of possible active motifs without time-consuming steps.

Computational study of Mg insertion into amorphous silicon: advantageous energetics over crystalline silicon for Mg storage

2013 ◽  
Vol 1540 ◽  
Author(s):  
Fleur Legrain ◽  
Oleksandr I. Malyi ◽  
Teck L. Tan ◽  
Sergei Manzhos
Keyword(s):  
Density Functional ◽  
Nearest Neighbor ◽  
Defect Formation ◽  
Crystalline Silicon ◽  
Computational Study ◽  
Binding Energies ◽  
Functional Theory ◽  
Wide Range ◽  
Insertion Sites ◽  
Formation Energies

ABSTRACTWe show in a theoretical density functional theory study that amorphous Si (a-Si) has more favorable energetics for Mg storage compared to crystalline Si (c-Si). Specifically, Mg and Li insertion is compared in a model a-Si simulation cell. Multiple sites for Mg insertion with a wide range of binding energies are identified. For many sites, Mg defect formation energies are negative, whereas they are positive in c-Si. Moreover, while clustering in c-Si destabilizes the insertion sites (by about 0.1/0.2 eV per atom for nearest-neighbor Li/Mg), it is found to stabilize some of the insertion sites for both Li (by up to 0.27 eV) and Mg (by up to 0.35 eV) in a-Si. This could have significant implications on the performance of Si anodes in Mg batteries.

First-principles study of the effect of lanthanum on the niobium diffusion in fcc Fe

2016 ◽  
Vol 30 (23) ◽  
pp. 1650157
Author(s):  
Xueyun Gao ◽  
Huiping Ren ◽  
Chunlong Li ◽  
Haiyan Wang ◽  
Huijie Tan
Keyword(s):  
First Principles ◽  
Density Functional ◽  
Nearest Neighbor ◽  
Transformation Process ◽  
Binding Energies ◽  
Diffusion Activation Energy ◽  
Repulsive Interaction ◽  
Functional Theory ◽  
The Density Functional Theory ◽  
And Migration

The effect of La on the diffusion of Nb in fcc Fe has been investigated using the first-principles calculations based on the density functional theory. The binding energies of Nb–vacancy, La–vacancy and La–Nb pairs have been calculated. The interactions of Nb–vacancy and La–Nb are attractive in 1nn and 2nn configurations (nn: nearest–neighbor). La atom attracts strongly with the 1nn vacancy, but has a weakly repulsive interaction with the 2nn vacancy. We consider four different Nb jumps in the presence of La atom to investigate the Nb diffusion in terms of vacancy formation and migration energy. The results suggest that La increases the diffusion activation energy of Nb in fcc Fe matrix, and is helpful to decelerate the Nb-diffusion-involved phase transformation process.

Sign in / Sign up

Export Citation Format

Share Document