Exploring the Influence of Applied Potential and Catalyst Morphology on the Selectivities of Copper Toward Carbon Dioxide Electroreduction Using Machine Learning
Исследование влияния приложенного потенциала и морфологии катализатора на селективность меди в отношении электрохимического восстановления диоксида углерода с использованием машинного обучения
2025-06-23
SCID: 54.1/ez67gnvg
Discuss with AI
Artificial neural networksCO2 electroreductionCatalyst morphologyCopper catalystsSHAP analysis
Figures from the paper
Abstract (AI)
Electrochemical CO 2 reduction (CO 2 RR) presents a sustainable approach to converting greenhouse gases into valuable chemicals. However, achieving precise product selectivity remains challenging. This study investigates the influence of applied potential and catalyst morphology on CO 2 RR selectivity using a machine learning (ML) framework. Unlike most existing ML applications which primarily rely on open databases, we collect 378 experimental data points from literature focusing on Cu and Cu‐derived catalysts in H‐type cells. Various ML models are trained using both numerical and categorical parameters, with the artificial neural network model demonstrating the highest accuracy in predicting Faradaic efficiencies of key CO 2 RR products. The SHapley Additive exPlanations (SHAP) analysis identifies applied potential as the primary determinant of product distribution. Additionally, it reveals synergistic effects between catalyst morphology and size, influencing pathways toward CO, C 1 , and C 2+ products. These insights offer guidance for the rational design of Cu‐based CO 2 RR systems by optimizing operational parameters to enhance multi‐carbon product selectivity. The study introduces a novel dataset and a data‐driven methodology that integrates experimental findings with ML techniques, advancing the mechanistic understanding of CO 2 RR and providing a scalable approach for catalyst discovery and optimization.
Key Findings
1
Among evaluated machine-learning models using numerical and categorical features, the artificial neural network achieves the highest accuracy for predicting Faradaic efficiencies of major CO₂ reduction products.
2
Catalyst morphology and size exhibit synergistic effects that influence selectivity toward CO, C1, and C2+ products.
3
SHAP analysis identifies applied potential as the primary factor governing product distribution during Cu-catalyzed CO₂ electroreduction.
4
The integrated experimental dataset and machine-learning framework support rational optimization of operational conditions and Cu catalyst structure for enhanced multicarbon-product selectivity.
5
The study compiles a novel dataset of 378 literature-derived experimental measurements for Cu and Cu-derived catalysts tested in H-type CO₂ reduction cells.
Research Object
Cu and Cu-derived catalysts for electrochemical CO2 reduction in H-type cells
Research Subject
The effects of applied potential and catalyst morphology and size on CO2 reduction product selectivity, including Faradaic efficiencies and pathways to CO, C1, and C2+ products
Publication Details
Publication Date
2025-06-23
Journal
Publisher
ISSN
Cited by
21
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai3
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation2013
Double-slit photoelectron interference in strong-field ionization of the neon dimer2018
Machine-Learning-Accelerated Catalytic Activity Predictions of Transition Metal Phthalocyanine Dual-Metal-Site Catalysts for CO2 Reduction2021