Complex ligand adsorption on 3D atomic surfaces of synthesized nanoparticles investigated by machine-learning accelerated ab initio calculation

Nanoparticle surfaces are passivated by surface-bound ligands, and their adsorption on synthesized nanoparticles is complicated because of the intricate and low-symmetry surface structures. Thus, it is challenging to precisely investigate ligand adsorption on synthesized nanoparticles. Here, we applied machine-learning-accelerated ab initio calculation to experimentally resolved 3D atomic structures of Pt nanoparticles to analyze the complex adsorption behavior of polyvinylpyrrolidone (PVP) ligands on synthesized nanoparticles. Different angular configurations of large-sized ligands are thoroughly investigated to understand the adsorption behavior on various surface-exposed atoms with intrinsic low-symmetry. It is revealed that the ligand binding energy (E-ads) of the large-sized ligand shows a weak positive relationship with the generalized coordination number((CN)) . This is because the strong positive relationship of short-range direct bonding (E-bind) is attenuated by the negative relationship of long-range van der Waals interaction (E-vdW). In addition, it is demonstrated that the PVP ligands prefer to adsorb where the long-range vdW interaction with the surrounding surface structure is maximized. Our results highlight the significant contribution of vdW interactions and the importance of the local geometry of surface atoms to the adsorption behavior of large-sized ligands on synthesized nanoparticle surfaces.

Automated summary

The study focuses on the complex adsorption behavior of polyvinylpyrrolidone (PVP) ligands on synthesized Pt nanoparticles. Machine-learning-accelerated ab initio calculation is applied to analyze the experimentally resolved 3D atomic structures. The results reveal the weak positive relationship between the binding energy of large-sized ligand and the generalized coordination number, as well as the preference of PVP ligands to adsorb where long-range van der Waals interaction is maximized.