Related Experiment Video
Updated: Jun 27, 2025
07:30
Rapid Nanoprobe Signal Enhancement by In Situ Gold Nanoparticle Synthesis
Published on: March 7, 2018
7.6K
Machine Learning-Accelerated High-Throughput Computational Screening: Unveiling Bimetallic Nanoparticles with
Kaiwei Wan1,2, Hui Wang1,2, Xinghua Shi1,2
1Laboratory of Theoretical and Computational Nanoscience, National Center for Nanoscience and Technology, Chinese Academy of Sciences, Beijing 100190, China.
ACS Nano
|May 2, 2024
Summary
We developed a machine learning approach to predict the peroxidase-like (POD-like) activity of bimetallic nanoparticles. This method accelerates the discovery of highly effective catalysts for various applications.
Area of Science:
- * Nanotechnology and Materials Science
- * Computational Chemistry
- * Catalysis
Background:
- * Bimetallic nanoparticles (NPs) exhibit crucial peroxidase-like (POD-like) activity for applications in biosensing, medicine, and environmental remediation.
- * Developing structure-activity relationships for bimetallic NPs is challenging due to numerous tunable properties.
- * Existing methods struggle to efficiently discover novel, high-performance bimetallic catalysts.
Purpose of the Study:
- * To establish robust scaling relationships for predicting POD-like activity in bimetallic NPs.
- * To develop a versatile descriptor for bimetallic NPs.
- * To accelerate the discovery of efficient bimetallic catalysts using machine learning.
Main Methods:
- * Analysis of catalytic reaction networks in pure metal NPs to establish scaling relationships.
- * Development of a novel descriptor for bimetallic NPs based on adsorption energy.
- * Integration of the descriptor into a machine learning-accelerated high-throughput computational workflow.
- * Prediction of POD-like activity for 1260 bimetallic NPs.
Main Results:
- * Established volcano-shaped correlations between catalytic activity and O* adsorption energy.
- * Achieved significantly boosted predictive accuracy for POD-like activity.
- * Identified several highly effective bimetallic NP catalysts.
- * Successfully predicted activities for a large number of bimetallic NP candidates.
Conclusions:
- * The developed machine learning approach effectively predicts POD-like activity of bimetallic NPs.
- * The study provides strategies for designing efficient bimetallic NP catalysts.
- * This work accelerates the discovery and design of advanced nanomaterials for various applications.
