Hydrophilic interaction chromatography in nonaqueous elution mode for separation of hydrophilic analytes on silica-based packings with noncharged polar bondings

Chromatographic effects of dedicated stationary and mobile phase variations in hydrophilic interaction chromatography (HILIC) were investigated using a set of nucleobases, nucleosides and deoxynucleosides as polar test solutes. Retention and selectivity profiles were comparatively mapped on four in-house developed silica materials modified with short alkyl chains (C-4, C-5) which carry hydroxyl functionalities (including diol motifs) as well as embedded sulphide or sulphoxide groups. These data were complemented by results obtained with two commercially available diol-type phases and a bare silica column. Besides elucidation of packing-related aspects this work concentrated specifically on extending aqueous HILIC (AQ-HILIC) to nonaqueous polar-organic elution conditions herein termed NA-HILIC. The exchange of the polar modifier water by various alcohols in ACN-rich mobile phases containing 5 mM ammonium acetate decreased the eluotropic strength of the resulting eluents. The gain in retention largely followed the order ethanol (EtOH) > methanol (MeOH) > 1,2-ethanediol (Et(OH)(2) and was accompanied by distinct effects on chromatographic selectivity. For example, on the most polar home-made packin'a the purine nucleoside selectivity guanosine/adenosine increased from 2.25 in the AQ-HILIC (k(guanosine) = 8.3) to 7.33 (k(guanosine) = 59) in the NA-HILIC mode when EtOH was employed as NA modifier while this value was 5.84 and 2.93 with MeOH and Et(OH)2, respectively (eluent: 5 mM ammonium acetate in ACN/modifier 90:10 v/v). Besides the type of protic modifier its percentage as well the retention and selectivity effects upon varying the ammonium acetate concentration and column temperature, respectively, were also investigated. Notable inter-column differences were found for all of these elution parameters. A mixed-mode retention model composed of partitioning and adsorption is proposed for both AQ- and NA-HILIC retention processes. The potential of (i) the implementation of novel polar bondings (such as ones containing sulphoxide functionalities) and (ii) the comprehensive exploitation of elution variables (type of protic modifiers, salt, ctc.) for providing new selectivity increments to the separation of polar analytes in HILIC is emphasised.