Machine learning aided design of single-atom alloy catalysts for methane cracking

Area of Science:

• Materials Science
• Catalysis
• Chemical Engineering

Background:

• Methane cracking is crucial for hydrogen production.
• Traditional methods face deactivation from carbon deposition.
• Developing stable and efficient catalysts is essential.

Purpose of the Study:

• To discover effective catalysts for methane cracking at low temperatures.
• To overcome catalyst deactivation issues in methane conversion.
• To explore single-atom alloy catalysis for hydrogen production.

Main Methods:

• Screening 10,950 transition metal single-atom alloy surfaces using machine learning.
• Predicting C-H dissociation energy barriers.
• Experimental validation of promising candidates like Re/Ni and Ir/Ni.

Main Results:

• Re/Ni single-atom alloy demonstrates high performance in methane cracking.
• Achieved 10.7 gH2 gcat-1 h-1 yield, 99.9% selectivity, and 7.75% CH4 conversion at 450°C.
• Sustained methane cracking for over 240 hours, surpassing existing methods.

Conclusions:

• Re/Ni single-atom alloys are effective catalysts for methane cracking.
• Mechanical energy input enhances methane conversion efficiency.
• This method offers a stable and efficient pathway for hydrogen production.