(665f) Dynamic Mean Field Theory for Fluids in Porous Materials: Comparison with Higher Order Approximations and Molecular Simulations

Recently we have developed a dynamic mean field theory (DMFT) for fluids in porous materials (P. A. Monson, J. Chem. Phys., 128, 084701 (2008)). The theory can be used to described the relaxation processes in the approach to equilibrium or metastable states for fluids in pores and is especially useful for studying system exhibiting adsorption/desorption hysteresis. The present paper deals with assessing the accuracy of this new approach to the dynamics of confined fluids. We do this in two ways. First we have investigated a higher order approximation to the dynamics by making use of the path probability method (PPM). This yields a dynamic theory that in the limit of long times yields the thermodynamic description of he system equivalent to the Bethe-Peierls or Quasi-Chemical approximation. Second we have carried out Monte Carlo simulations using Kawasakii dynamics. Both the DMFT and PPM can be regarded as approximations to the results averaged over an ensemble of Kawasaki dynamics simulations.

We have studied the dynamics of uptake processes in a slit pore geometry with the DMFT, the PPM and Monte Carlo simulations for a range of temperatures and slit widths in both the complete wetting case and the partial wetting case. We show that for temperature sufficiently lower than the critical point a) the mechanisms of uptake processes as determine in Monte Carlo simulations are correctly predicted by the DMFT and PPM. The PPM offers no new insights as far as qualitative picture is concerned and is an order of magnitude higher in computational expense than the DMFT.