Pulsed field gradient nuclear magnetic resonance has been successfully applied to a direct and detailed experimental study of topological and dynamic aspects involved in the exchange of small, nonsorbed fluid molecules between the intraparticle pore network and the interparticle void space in chromatographic columns packed with spherical-shaped, porous particles. The approach provides quantitative data about the effective, intraparticle diffusion coefficients (and tortuosity factors) and about the associated, diffusion-limited mass transfer kinetics, including stagnant boundary layer contributions. In view of the recorded exchange kinetics, an analytical description for solute diffusion into/out of spherical particles is offered and addresses the influence of the particle size distribution and particle shape on the observed mass transfer rates and calculated diffusivities. The combined analyses of the steady-state intraparticle pore diffusion data and the associated exchange kinetics with Peclet numbers up to 500 reveals the existence of external stagnant fluid where all the interparticle fluid-side resistance to diffusion is localized. It is represented by a thin stagnant boundary layer around the particles and can be accounted for by the introduction of a hydrodynamically effective particle diameter which is found to depend on the Peclet number. The approach appears to be promising for a selective, detailed study of the boundary layer dynamics. Concerning the investigation of different chromatographic media and intraparticle morphologies, we demonstrate that the actual correlation (or randomness) of interconnection between intraparticle pores of different size has a profound effect on the observed tortuosity factors and the diffusion-limited stagnant mobile phase mass transfer kinetics. Compared to intraparticle pore networks with a random assignment of different pore sizes, hierarchically structured bidisperse porous particles offer a superior network topology, which can form the basis for an increased chromatographic performance.