Partitioning of spherical particles into fibrous matrices

Partition coefficients are computed for hard spherical solutes in equilibrium with a three-dimensional network of random fibers using grand canonical Monte Carlo simulations. Spherical boundary conditions are employed, instead of the traditional toroidal conditions, to eliminate inhomogeneities within the fibrous matrix. The partition coefficients obtained by Monte Carlo simulation compare favorably at low matrix densities with those estimated in the first paper of this series using density-functional theory coupled with a superposition approximation. A rational approximant is constructed which describes the simulation results for all densities and solute concentrations.