Stepping Out of Transition Metals: Activating the Dual Atomic Catalyst through Main Group Elements

Abstract

In recent years, investigations into atomic catalysts has accelerated significantly. Although different atomic catalysts have been developed, the introduction of main group elements is rarely considered. In this work, the possibility of introducing alkaline/alkaline earth metals (AAEM), post‐transition metal (Post‐TM), and metalloids to form stable graphdiyne‐based dual atomic catalysts (GDY‐DAC) is revealed. The main group elements not only act as a promising separator to improve the loading of DACs but also activate the alkyl chains to facilitate the electroactivity of GDY‐DAC. Most importantly, the main group elements in the GDY‐DAC do not affect the electroactivity of transition or lanthanide metals and even enable subtle modulations on the electronic structures. The p band center is a significant descriptor to modulate the electroactivity in oxides while their applications in the atomic catalysts are unclear. With the further evaluations of machine learning, it is found that the involvements of s‐orbitals and p‐orbitals perturb the prediction accuracies of both formation energies and p‐band center, especially for the AAEM. This work supplies insights that are expected to aid progress in designing main group elements‐based atomic catalysts, which opens a new avenue in designing advanced electrocatalysts.

Entanglement of Spatial and Energy Segmentation for C1 Pathways in CO2 Reduction on Carbon Skeleton Supported Atomic Catalysts

“…Furthermore, as a powerful tool, the machine learning (ML) technique has been introduced to predict the thermodynamic properties of electrocatalysts. [53][54][55] In this work, we have applied the Gaussian Process Regression (GPR) algorithm, which is able to obtain predictions based on a few parameters. Such methods are highly useful for the relatively smaller datasets in our work when compared with many state-of-the-art machine learning models.…”

Application of machine learning for advanced material prediction and design

“…The increased deviations of the predictions suggest that the electronic structures cannot be fully revealed by the physicochemical properties. Recently, they have further extended the research from transition metal and lanthanide metals to the main group elements 126 . Based on the similar GPR method, they have identified that the involvements of s and p orbitals have evidently perturbed the prediction accuracies of both formation energies and p‐band center, especially the alkaline/alkaline earth metals, which show much higher RMSE than other groups (Figure 11G–I).…”

“…Copyright 2021 Wiley‐VCH). The comparison of the GPR model predicted data and original data of the formation energy in (G) alkaline/alkaline earth metals based GDY‐DAC, (H) post‐transition metals based GDY‐DAC and (I) metalloid based GDY‐DAC (Reproduced with permission 126 . Copyright 2021 Wiley‐VCH)…”

Low‐dimensional material supported single‐atom catalysts for electrochemical CO2 reduction

“…They further experimented with verifying that a Zn/Ln DMS catalyst was highly effective for ECR to produce CO/H 2 syngas with a tunable ratio 159 . In addition, using the same machine learning techniques, they also discover that combining main group elements with TM and Ln metals is able to form the stable graphdiyne‐based DMS with high electroactivity 161 …”

“…159 In addition, using the same machine learning techniques, they also discover that combining main group elements with TM and Ln metals is able to form the stable graphdiyne-based DMS with high electroactivity. 161…”