The mass transfer mechanism in HILIC was investigated in depth. The reduced heights equivalent to a theoretical plate (HETP) of five low molecular weigh compounds with retention factors of −0.05 (slight exclusion from the surface due to the presence of a water-rich layer in which naphthalene is insoluble) to 3.64 were measured at room temperature for a 4.6mm×100mm column packed with 3.5μm 140Å XBridge HILIC particles in a wide range of flow velocities. The mobile phase was a buffered acetonitrile–water mixture (92.5/7.5, v/v). Using a physically reliable model of effective diffusion in binary composite media (Torquato's model), the longitudinal diffusion and solid–liquid mass transfer resistance reduced HETP terms were measured. The reduced short-range eddy dispersion HETP was taken from the literature data. The long-range reduced HETP was directly measured from the subtraction of these HETP terms to the overall HETP measured from moment analysis. In contrast to RPLC, the plots of the reduced HETP versus the reduced velocity depend weakly on the retention factor, due to the constant, low intra-particle diffusivity observed in HILIC. So, the reduced longitudinal diffusion HETP is smaller and the reduced solid–liquid mass transfer resistance HETP is larger in HILIC than in RPLC. Whereas border effects can be concealed in RPLC for retained analytes due to fast radial equilibration across the column diameter, a residual long-range eddy dispersion term persists in 4.6mm I.D. HILIC columns, even at very slow flow rates. Experiments show that the minor differences in the long-range eddy dispersion term between analytes having different retention factors is directly correlated to the reciprocal of their bulk diffusion coefficient. The performance of HILIC columns packed with fine particles is then more sensitive to the inlet sample distribution and to the outlet sample collection than RPLC columns due to the relatively poor radial mixing controlled by lateral diffusion.
Read full abstract