Computational Prediction of Ligand Curvature-Selectivity Ability on Gold Nanoparticle Surface

<br/>Organic ligands are foundational in determining the properties and assembly of inorganic nanoparticles. Recent works have focused on ligands that passivate the nanocrystal surface to enable the supramolecular structure to be formed. However, precise nanoparticle surface modification, as a means of controlling the nanoparticle assembly structure, is difficult. Therefore, in this work, we devised a ligand (CNAr^Mes2) that can achieve site selective binding on gold nanoparticle (AuNP) surfaces, based on curvature. We adopted a variety of computational approaches, namely quantum mechanical electronic structure theory calculations and molecular dynamics (MD) simulations, to obtain microscopic insights into the AnNP/ligand interaction energies, dynamics, solvation properties, and spectroscopy. We were able to compute the ligand adsorption characteristic and further predict the ligand adsorption sites on AuNPs, as quantified by four unique descriptors: the site binding ratio, the edge-binding factor, and the ligand solubility and solvent partition factors. We verified our predictions experimentally using liquid-liquid extraction, transmission electron microscopy, zeta potential measurements and surface-enhanced Raman spectroscopy. Together, our investigations indicate that CNAr^Mes2 provided good selectivity for AuNPs based on their average sizes, with a separation resolution of ± 5Å that is competitive with the state of the art in industry. Our study demonstrates how first-principles based computations can guide experimental design at the nanoscale, to enable novel material physics and commercial processes.