Related Experiment Video
Updated: Jul 15, 2026
05:47
Preparation of Polyoxometalate-based Photo-responsive Membranes for the Photo-activation of Manganese Oxide Catalysts
Published on: August 7, 2018
Machine learning-optimized bioinspired N-doped carbon-wrapped trimetallic oxides for an efficient oxygen evolution
Farhan Zafar1, Hamdy Khamees Thabet2, Muhammad Asad3
1Department of Chemistry, COMSATS University Islamabad, Lahore Campus, Lahore 54000, Pakistan. sadafulhassan@cuilahore.edu.pk.
Nanoscale
|July 14, 2026
Summary
Machine learning accelerates the discovery of efficient electrocatalysts for the oxygen evolution reaction (OER). This study optimized trimetallic oxides using ML, identifying a purine-derived catalyst with superior performance.
Area of Science:
- Materials Science
- Electrochemistry
- Computational Chemistry
Background:
- Developing efficient electrocatalysts for the oxygen evolution reaction (OER) is crucial for sustainable energy technologies.
- Nitrogen-doped multi-metal oxides show promise, but understanding synergistic effects of metal nodes and nitrogen sources is challenging.
- Conventional experimental methods are inefficient for optimizing complex multi-metallic catalyst systems.
Purpose of the Study:
- To employ machine learning (ML) for optimizing nitrogen-doped multi-metal oxide electrocatalysts for OER.
- To elucidate the individual contributions of metal nodes (Fe, Co, Zn) and nitrogen sources in OER performance.
- To accelerate the screening and design of highly efficient OER electrocatalysts.
Main Methods:
- Fabrication of trimetallic FeCoZn squarate metal-organic frameworks (FCZ-Sq MOFs).
- Two-stage ML optimization to determine optimal metal ratios and select effective nitrogen sources (purine, pyridine, xanthine).
- Calcination of ML-optimized materials to form N-doped carbon-coated trimetallic oxides (NC@FCZ-Ox).
Main Results:
- The purine-derived catalyst (NCPu@FCZ-Ox) exhibited superior OER activity: low overpotential (270 mV at 10 mA cm-2), low onset potential (1.40 V vs. RHE), and a Tafel slope of 74 mV dec-1.
- ML models successfully identified optimal metal ratios and the most effective nitrogen source, significantly outperforming other catalysts and pristine MOFs.
- The study provides fundamental insights into structure-performance relationships in multi-metallic and N-doped systems.
Conclusions:
- Machine learning is a powerful tool for accelerating the discovery and optimization of advanced electrocatalysts.
- The ML-optimized NCPu@FCZ-Ox catalyst demonstrates high efficiency for the oxygen evolution reaction.
- This work pioneers the use of ML for precisely optimizing metal nodes and nitrogen sources in catalyst design for sustainable energy applications.
