(105c) The Role of Chloride in the Anisotropic Growth of Cu Nanocrystals: A DFT Study

A significant challenge in the development of functional nanomaterials is understanding the growth of colloidal metal nanocrystals. There are many applications for metal nanocrystals, ranging from catalysts to various plasmonic and electronic devices. It is evident that the properties of the nanocrystals are exquisitely sensitive to their shapes. Halides have played a significant role in achieving shape control, but the exact mechanisms remain unknown. Our previous study revealed that synergistic interactions between hexadecylamine (HDA) capping agent and solution-phase chloride can lead to the anisotropic growth of the Cu seeds into nanowires with high aspect ratios. In this study, we hypothesized that adsorbed chloride on Cu surfaces would affect both the binding and diffusion of Cu atoms to effect shape control. We used quantum density-functional theory (DFT) calculations to show that 1/3 and 1/2 monolayer (ML) coverage of Cl on Cu(111) and 1/2 ML of Cl on Cu(100) are the configurations with the lowest surface energies as a function of the solution-phase concentration (chemical potential) of chloride, consistent with experiment. When the Cu surfaces contain adsorbed chloride, the surface energy of Cl-Cu(111) is higher than Cl-Cu(100) and a stronger binding of Cu adatoms on Cl-Cu(111) is observed. This is the opposite trend to what occurs on the bare Cu surfaces. We also find that absorbed Cl facilitates the diffusion of Cu adatoms on Cu(100) by decreasing the diffusion-energy barrier from 0.55 eV on bare Cu(100) to 0.15 eV on 0.5 ML Cl-Cu(100), but hinders the diffusion on Cu(111) by increasing the barrier from 0.04 eV on the bare surface to 0.23 eV on the Cl-covered surface. Furthermore, the {100}-{111} interfacet barrier favors transport from Cu(100) to Cu(111). All of these factors promote the growth of Cu into nanowires with long {100} side facets and {111} end facets. A model based on the theory of absorbing Markov chains indicates that nanowires with high aspect ratios, comparable to those in experiment, are predicted based on values of DFT-predicted surface diffusion coefficients.