Amorphous MoS2 from a machine learning inter-atomic potential

Area of Science:

• Materials Science
• Catalysis
• Computational Chemistry

Background:

• Amorphous molybdenum disulfide (MoS2) is a promising electrocatalyst for hydrogen evolution.
• The atomic structure and origin of its high activity remain poorly understood.

Purpose of the Study:

• To elucidate the atomic structure of amorphous MoS2.
• To understand the relationship between structure and electrocatalytic activity.

Main Methods:

• Developed a classical inter-atomic potential using charge equilibration neural network.
• Generated amorphous MoS2 models via melting and quenching.
• Performed density functional theory (DFT) calculations.

Main Results:

• Identified low-coordinated Mo (4- and 5-coordinated) and S-S bonds in amorphous MoS2.
• DFT revealed metallic character with significant 4-coordinated Mo contribution at the Fermi level.
• Observed S-S bond formation linked to sulfur reduction and broadened 3s states.

Conclusions:

• The unique atomic structure, including low-coordination Mo and S-S bonds, contributes to the metallic nature and high electronic density of states.
• These characteristics are likely responsible for the enhanced electrocatalytic activity of amorphous MoS2 for hydrogen evolution.