Direct discovery · Chemistry · Materials science

Active machine learning helps identify CO2 electrocatalysts

Cu-Al electrocatalysts for CO2-to-ethylene conversion identified through active machine learning and experiments.

Summary

The Nature paper reports Cu-Al electrocatalysts identified through density functional theory calculations combined with active machine learning. The catalysts reduced CO2 to ethylene with reported Faradaic efficiency above 80% under the tested conditions.

AI role

Prioritized electrocatalyst candidates by combining DFT calculations with active machine learning.

Narrative role

This event adds energy chemistry to the acceleration story, where AI narrows candidates before experimental characterization.

Caveat

Performance is condition-specific and does not by itself establish industrial catalytic viability.