Authors: Haoyu Yang, Juanli Zhao, Qiankun Wang, Bin Liu, Wei Luo, Ziqi Sun, Ting Liao

Published on: February 06, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.03876

Summary

• What is new: Integrating multi-branch Convolutional Neural Network (CNN) analysis models with hybrid descriptor based activity volcano plots for screening efficient Single-atom catalysts (SACs) for CO2 reduction.
• Why this is important: Difficulty in finding optimal SAC systems for catalyzing chemical reactions due to diverse combinations of active elements and support materials.
• What the research proposes: A deep learning approach using a multi-branch CNN model to predict adsorption energies and a hybrid-descriptor for constructing volcano plots, enabling intuitive screening of catalyst candidates.
• Results: Developed a framework that can effectively screen for efficient SACs, providing insights into the electronic and elementary descriptors essential for catalytic activity.

Technical Details

Technological frameworks used: Multi-branch Convolutional Neural Network (CNN)

Models used: Deep learning models for predicting adsorption energies

Data used: 2D electronic density of states

Potential Impact

Chemical manufacturing industries, particularly those involved in CO2 reduction and sustainable energy solutions.

Want to implement this idea in a business?

We have generated a startup concept here: CataLeap.