Practical observations on the performance of bare silica in hydrophilic interaction compared with C18 reversed-phase liquid chromatography.

The kinetic performance of a bare silica and C18 phase prepared from the same sub-2μm and 3.5μm base materials were compared in the HILIC and RP mode using both charged and neutral solutes. The HILIC column was characterised using the neutral solute 5-hydroxymethyluridine, the weak base cytosine, and the strong base nortriptyline, the latter having sufficient retention also in the RP mode to allow comparison of performance. Naphthalene was also used as a simple neutral substance to evaluate the RP column alone. The retention factors of all substances were adjusted to give similar values (k'∼5.5) at their respective optimum linear velocities. Reduced van Deemter b-coefficients (determined by curve fitting and by the peak parking method, using a novel procedure involving switching to a dummy column) were significantly lower in HILIC for all substances compared with those found under RP conditions. Against expectation, c-coefficients were always lower in RP when compared with HILIC using sub-2μm particles. While measurement of these coefficients is complicated by retention shifts caused by the influence of high pressure and by frictional heating effects, broadly similar results were obtained on larger particle (3.5μm) phases. The mechanism of the separations was further investigated by examining the effect of buffer concentration on retention. It was concluded that HILIC can sometimes show somewhat inferior performance to RP for fast analysis at high mobile phase velocity, but clearly shows advantages when high column efficiencies, using longer columns at low flow velocity, are employed. The latter result is attributable to the lower viscosity of the mobile phase in HILIC and the reduced pressure requirement as well as the lower b-coefficients.