Morphology-transport relationships in liquid chromatography: Application to method development in size exclusion...

We present relationships between the multiscale structure and the separation properties of size exclusion chromatography (SEC) columns. Physical bed reconstructions of wall and bulk regions from a 2.1 mm i.d. column packed with fully porous 1.7 µm bridged-ethyl hybrid (BEH) particles, obtained by focused ion-beam scanning electron microscopy, serve as geometrical models for the packing microstructure in wall and central regions of a typical narrow-bore SEC column. In addition, the intraparticle mesopore space morphology of the BEH particles is reconstructed using electron tomography, to ultimately construct a realistic multiscale model of the bed morphology from mesopore level via interparticle macropore space to transcolumn scale. Complemented by the results of eddy dispersion simulations in computer-generated bulk packings, relationships between packing microstructure and transchannel, short-range interchannel, as well as transcolumn eddy dispersion are used to analyze the fluid dynamics in the interparticle macropore space of the model. Further, we simulate hindered diffusion and accessible porosity for passive, finite-size tracers in the intraparticle mesopore space, to finally determine the effective particle and bed diffusion coefficients of these tracers in the hierarchical (macro-mesoporous) bed. Retention and transport properties of polystyrene standards with hydrodynamic diameters from 5 to 95 Å in tetrahydrofuran are subsequently predicted without introducing bias from arbitrary models. These properties include the elution volumes of the polystyrene standards, the global peak capacity (over the entire separation window), and the rate of peak capacity at any fixed elution volume. Optimal flow rates yielding maximal global peak capacity and a nearly uniform rate of peak capacity over the entire separation window are close to 0.04 and 0.20 mL/min, respectively. SEC column performance obtained for fully porous and superficially porous particles is compared by varying the core-to-particle diameter ratio ρ from 0 to 0.95. Because the separation window is narrowing more rapidly than the rate of peak capacity is growing with increasing ρ, core-shell particles always provide smaller global peak capacity; they still can be advantageous but only for simple sample mixtures. The presented morphology-performance approach holds great promise for method development in SEC.