Structural evolution in gold nanoparticles using artificial neural network based interatomic potentials.

Area of Science:

• Computational materials science
• Atomic and molecular physics
• Nanotechnology

Background:

• Relativistic effects significantly alter gold's chemical and physical properties compared to other metals.
• Gold's unique behavior means it does not necessitate perfectly symmetrical structures for stability.

Purpose of the Study:

• To investigate the structural evolution of gold clusters into nanoparticles.
• To identify stable structural configurations across various cluster sizes.
• To understand the transition from amorphous to symmetric structures as gold nanoparticles grow.

Main Methods:

• Utilized an artificial neural network interatomic potential trained on quantum mechanical data.
• Explored the potential energy surface for "magic" number gold clusters (Au309, Au561, Au923).
• Analyzed the stability of symmetric versus distorted symmetric core structures with amorphous outer layers.

Main Results:

• Identified distorted symmetric cores with amorphous atomic layers as more stable for medium-sized gold clusters.
• Discovered new potential magic numbers (Au160, Au327, Au571) exhibiting stable symmetric cores.
• Observed a trend towards more symmetric core structures in larger nanoparticles (Au923, <3.3 nm).

Conclusions:

• Gold nanoparticle stability is governed by a transition from amorphous layers to increasingly symmetric cores with size.
• The findings challenge the assumption of complete symmetry in stable gold nanostructures.
• New magic numbers and structural evolution pathways for gold nanoparticles have been elucidated.