Evolution of Metastable Structures at Bimetallic Surfaces from Microscopy and Machine-Learning Molecular Dynamics

Area of Science:

• Materials Science
• Surface Science
• Heterogeneous Catalysis

Background:

• Surface composition and morphology in bimetallic systems often differ significantly from bulk properties.
• Understanding interface restructuring is critical for applications in catalysis and materials science.

Purpose of the Study:

• To provide an unprecedented atomistic view of the long-time scale restructuring of palladium (Pd) deposited on silver (Ag).
• To identify and characterize novel surface restructuring mechanisms in unbiased manner.

Main Methods:

• Utilized a combination of microscopy and spectroscopy.
• Developed and applied accelerated machine-learning molecular dynamics simulations.
• Employed an automated analysis method for mechanistic investigation.

Main Results:

• Discovered previously unidentified restructuring mechanisms: Pd-Ag place exchange, Ag pop-out, and step dynamics.
• Observed layer-by-layer Pd dissolution preceded by Ag encapsulation of Pd islands.
• Identified significant Ag migration and extensive vacancy pit formation within microseconds.
• Characterized metastable Ag-encapsulated Pd structures with enhanced reactant accessibility.

Conclusions:

• The study reveals novel, unbiased mechanisms of surface restructuring in Pd-Ag systems.
• Ag encapsulation of Pd islands creates catalytically important metastable structures.
• The developed computational approach is broadly applicable to complex multimetallic systems for mechanistic investigations.