As global efforts to combat climate change intensify, electrochemical CO₂ reduction reaction (CO₂RR) stands out as a critical technology for converting greenhouse gases into valuable fuels and chemicals. Yet, a persistent bottleneck has hindered progress: the lack of precise tools to characterize and optimize the porous catalyst layers where reactions occur. A new study published in Frontiers in Energy resolves this challenge by deploying deep learning (DL) to map microscopic structures and simulate mass transport in CO₂RR catalyst layers with unprecedented accuracy.

A research team at Shanghai Jiao Tong University has developed a systematic framework that combines semantic-segmentation AI models with experimental validation to analyze catalyst layers (CLs). Using silver nanoparticles as catalysts and Nafion ionomer as a binder, the team fabricated CLs with varying ionomer-to-catalyst (I/C) ratios (0.2, 0.4, and 0.6) to dissect how composition affects performance.

Key Findings

• AI Outperforms Traditional Methods: Among five DL models tested, DeepLabV3plus achieved the highest segmentation accuracy (91.29%), surpassing conventional thresholding techniques (72.35–77.42%). This enabled precise extraction of pore parameters like porosity and pore size distribution, validated against mercury intrusion porosimetry (MIP).
• Optimal I/C Ratio Identified: CLs with an I/C ratio of 0.2 exhibited a twofold increase in gas diffusion distance (672.88 nm vs. 321.38 nm for I/C = 0.6), enabling 89% Faradaic efficiency (FE) for CO production at 250 mA/cm²—the highest reported in the study.
• Structure-Performance Link: Excessive ionomer (I/C ≥ 0.4) clogged macropores, reducing CO₂ accessibility and FE by 12%. Simulations confirmed that lower ionomer content preserved open pore networks critical for long-range mass transport.

The integrated framework translates nanoscale morphology directly into device-level performance metrics, providing a scalable blueprint for industrial CO₂ electrolyzers. By minimizing ionomer loading to 0.2, catalyst layers can sustain high current densities without sacrificing selectivity or durability, moving carbon-neutral fuel production closer to market deployment.
DOI: 10.1007/s11708-025-1029-x