Monte Carlo Computer Simulation Of Liquid Restricted-Diffusion In Porous Media

Diffusion is the most essential movement form and an important transfer activity of molecules. It is responsible for all chemical reactions. NMR is presently the only available non-invasive method that provides information of molecular displacements in a spatial scale comparable to cell dimensions of biological systems. Time-dependent self-diffusion coefficient of fluid in porous media can be obtained by pulsed-field gradient nuclear magnetic resonance. The data can be used to obtain microstructural information of porous media, such as surface-to-volume ratio, reciprocal tortuosity, and pore radii. This information is important for understanding various kinds of transport processes in porous media like permeability and electrical conduction. In previous reports, the studies of self-diffusion behavior of liquid in porous media have mostly been focused on short diffusion time, and there are fewer researches on self-diffusion with surface relaxation.In this paper, self-diffusion behavior of liquid in porous media at long diffusion time without and with surface relaxation was studied. The main work is summarized as follows:1. A Monte Carlo method was established to simulate the liquid diffusion in restricted cylinders without surface relaxation. Simulation results validate the correctness of simulation method and theoretical equations. Microstructural information of porous media, such as surface-to-volume ratio, reciprocal tortuosity, and pore radii, was obtained based on the simulated data.2. Different diffusion models were established to simulate the liquid diffusion in porous media with surface relaxation. Simulation results are in good agreement with theoretical equations. The simulated data were used to obtain surface-to-volume ratio and pore radii of porous media. A method for measuring the pore radii of cylindrical porous media at long diffusion time was proposed.