The effects of size, shape, and surface composition on the diffusive behaviors of nanoparticles at/across water–oil interfaces via molecular dynamics simulations

We have employed molecular dynamics simulations to systematically investigate the effects of nanoparticles’ structural and chemical properties on their diffusive behaviors at/across the water–benzene interface. Four different nanoparticles were studied: modified hydrocarbon nanoparticles with a mean diameter of 1.2 nm (1.2HCPs), modified hydrocarbon nanoparticles with a mean diameter of 0.6 nm (0.6HCPs), single-walled carbon nanotubes (SWCNTs), and buckyballs. We found that the diffusion coefficients of 0.6 and 1.2HCP were larger than the corresponding values predicted using the Stokes–Einstein (SE) equation and attributed this deviation to the small particle size and the anisotropy of the interface system. In addition, the observed directional diffusive behaviors for various particles were well-correlated with the derivative of the potential of mean force (PMF), which might indicate an effective driving force for the particles along the direction perpendicular to the interface. We also found that nanoparticles with isotropic shape and uniform surface, e.g., buckyballs, tend to have smaller diffusion coefficients than those of nanoparticles with comparable dimensions but anisotropic shapes and non-uniform surface composition, e.g., SWCNT and 0.6HCP. One possible hypothesis for this behavior is that the “perfect” isotropic shape and uniform surface of buckyballs result in a better-defined “solvation shell” (i.e., a shell of solution molecules), which leads to a larger “effective radius” of the particle, and thus, a reduced diffusion coefficient.