Exploring Configurations of Nanocrystal Ligands Using Machine-Learned Force Fields.

Area of Science:

• Materials Science
• Computational Chemistry
• Nanotechnology

Background:

• Semiconducting nanocrystals with organic ligands are crucial for applications like light harvesting and sensing.
• Experimental structural data for these complex systems is limited, hindering computational modeling.
• Accurate modeling requires addressing challenges like atomic partial charges and diverse ligand conformations.

Purpose of the Study:

• To develop and apply a machine-learned force field for investigating the surface chemistry of lead sulfide (PbS) nanocrystals.
• To explore the passivation mechanisms of acetate ligands on PbS surfaces.
• To overcome limitations of traditional computational methods in modeling nanocrystal systems.

Main Methods:

• Developed a machine-learned force field trained on Density Functional Theory (DFT) data.
• Utilized the force field to simulate PbS nanocrystals with acetate ligands.
• Analyzed ligand geometries, surface passivation, and predicted ligand infrared (IR) spectra.

Main Results:

• Demonstrated that carboxylate ligands adopt diverse "tilted-bridge" and "bridge" geometries to passivate metal-rich PbS surfaces.
• Investigated the resulting ligand IR spectra, providing vibrational signatures of passivation.
• Circumvented the issue of assuming fixed atomic partial charges, improving model accuracy.

Conclusions:

• Machine-learned force fields offer a powerful approach for accurate computational modeling of complex nanocrystal systems.
• The study provides detailed insights into the surface chemistry and ligand behavior of PbS nanocrystals.
• This work paves the way for advanced computational studies of nanomaterials.