Multivariate Bayesian Optimization of CoO Nanoparticles for CO₂ Hydrogenation Catalysis

Area of Science:

• Catalysis
• Materials Science
• Chemical Engineering

Background:

• CO2 hydrogenation is crucial for renewable fuels and chemicals.
• Developing selective and robust catalysts remains a significant challenge.
• Cobalt oxide (CoO) catalysts show promise, but performance depends on crystal phase and morphology.

Purpose of the Study:

• To systematically control CoO nanoparticle synthesis for optimized catalytic performance.
• To map the synthetic design space for colloidal CoO nanoparticles.
• To optimize nanoparticles for multiple catalytically relevant features within a target crystalline phase.

Main Methods:

• Multivariate Bayesian optimization coupled with a data-driven classifier.
• Colloidal nanoparticle synthesis for precise control over CoO attributes.
• Characterization and assessment of CoO/SiO2 catalysts for CO2 hydrogenation.

Main Results:

• Optimized synthesis yielded small, phase-pure rock salt CoO nanoparticles with uniform size and shape.
• The optimized CoO/SiO2 catalyst demonstrated higher activity and ~98% CH4 selectivity for CO2 hydrogenation.
• Optimized catalysts showed enhanced stability against sintering and carbon occlusion.

Conclusions:

• Bayesian optimization effectively navigates complex nanoparticle synthesis for tailored catalyst design.
• Optimized CoO nanoparticles offer a promising pathway for efficient CO2 conversion to methane.
• Catalyst stability and surface coverage are critical factors for high performance in CO2 hydrogenation.