| Summary: | Ever increasing control over the shape and form of a material's nanoscale features provokes the pursuit of a detailed understanding for the main factors influencing fluid transport. It is sought to facilitate the intelligent design of novel materials used in membrane separation processes. In addition to a strong dependence on molecular mobility, mass transport is heavily influenced by thermodynamic effects. Isolating thermodynamic and mobility effects is useful to understand the significant driving forces for mass transport through porous materials and their selective characteristics. However, experimental techniques are limited in probing this behaviour at the nanometre scale. In response to experimental challenges, the present study makes extensive use of the ability of molecular simulations to reflect the molecular character of nanoscale diffusion and identify equilibrium and transport properties individually. First, this work investigates diffusive mass transport inside a planar slit pore focusing on the influence of solid-fluid interactions, pore width, and fluid density. The influence of solid-fluid interactions, in particular, have often been neglected in studies of mass transport in porous solids. The vast variety of functionalised nano-materials is virtually endless and has spurred interest in this area. Equilibrium simulations were employed to determine self- and collective diffusivities and Grand Canonical insertions were used for the determination of thermodynamic factors. In addition, this work showcases the implementation of a highly efficient Non-Equilibrium Molecular Dynamics (NEMD) method through which effective transport was studied. The method was used to determine effective diffusivities which incorporate thermodynamic effects, the dominating contribution to transport for dense fluids. It is well suited to observe effective fluid transport in confined spaces as opposed to measuring self-diffusion, a measure for single-particle mobility only. The method is effective in studying mass transport in model systems as well as more realistic, complex geometries. As a second exemplary case, gas permeation through an atomistically detailed model of a high free-volume polymer was simulated explicitly with the NEMD approach. In addition to determining permeability and solubility directly from NEMD simulations, the results also shed light on the permeation mechanism of the penetrant gases, suggesting a departure from the expected pore-hopping mechanism due to the considerable accessibility of permeation paths.
|
|---|
