Efficient Machine-Learning-Aided Screening of Hydrogen Adsorption on Bimetallic Nanoclusters

Abstract

Nanoclusters add an additional dimension in which to look for promising catalyst candidates, since catalytic activity of materials often changes at the nanoscale. However, the large search space of relevant atomic sites exacerbates the challenge for computational screening methods and requires the development of new techniques for efficient exploration. We present an automated workflow that systematically manages simulations from the generation of nanoclusters through the submission of production jobs, to the prediction of adsorption energies. The presented workflow was designed to screen nanoclusters of arbitrary shapes and size, but in this work the search was restricted to bimetallic icosahedral clusters and the adsorption was exemplified on the hydrogen evolution reaction. We demonstrate the efficient exploration of nanocluster configurations and screening of adsorption energies with the aid of machine learning. The results show that the maximum of the d -band Hilbert-transform ϵ u is correlated strongly with adsorption energies and could be a useful screening property accessible at the nanocluster level.

Toward Excellence of Electrocatalyst Design by Emerging Descriptor‐Oriented Machine Learning

“…[24b] Jiang et al have also showed that an adjusted coordination number can act as a general descriptor to bridge the structure and reactivity for the oxygen sites over transitionmetal oxides. [25] In addition, if the properties are derived from electron density, [26] they are referred to as electronic descriptors. These descriptors are usually obtained from electronic structure calculations, which are time-consuming owing to the first principles/ab initio calculations required.…”

“…In addition, if the properties are derived from electron density, [ 26 ] they are referred to as electronic descriptors. These descriptors are usually obtained from electronic structure calculations, which are time‐consuming owing to the first principles/ab initio calculations required.…”

Predicting hydration layers on surfaces using deep learning

“…This will likely require development of the descriptor used in the training process, taking into account a more detailed description of the local chemical character than offered by the elemental name alone. 70,71…”

Machine Learning for Electrocatalyst and Photocatalyst Design and Discovery

“…Manifold learning algorithms generate low-dimensional representations that retain neighborhood relationships in the original high-dimensional data, allowing visualization and modeling (Figure ). , The t-distributed stochastic neighbor embedding (t-SNE) algorithm has been used extensively for the visualization of high-dimensional data in two or three dimensions. ,, However, the large time complexity of this algorithm, and variations in data visualizations that are highly dependent on the selection of hyperparameters, limit its applications. An improved algorithm, uniform manifold approximation and projection (UMAP) is an efficient dimensional reduction technique for data visualization and nonlinear dimension reduction .…”