Reversed-Phase and Hydrophilic Interaction Chromatography Columns for Size-Exclusion Chromatography

Anna Maria Caltabiano

doi:10.17918/etd-7632

Reversed-phase (RP) and hydrophilic interaction chromatography (HILIC) columns are very attractive for size-exclusion chromatography (SEC) separations because they are highly efficient, robust and rugged, and can provide short analysis times. These columns are mechanically strong, compatible with all organic solvents and water, stable at low and moderately high pH, require short equilibration times, and are relatively inexpensive. The feasibility of using common RP (e.g., C18, C8, C4, phenyl-, fluoro-phenyl-, and cyano-modified) and HILIC (e.g., diol-modified) columns in non-aqueous and aqueous SEC has been investigated. Successful SEC separations of synthetic polymers (e.g., polystyrene (PS), poly(methyl methacrylate) (PMMA), sodium polystyrene sulfonate (NaPSS)) were achieved on RP and HILIC columns, and of biopolymers on a HILIC column, with a variety of common solvents used as the mobile phase in high performance liquid chromatography (HPLC). A unique bimodal separation of solutes in a complex mixture on an RP (cyano-modified) column, such as hydrophobic interaction/retention of Vestipitant molecules (active pharmaceutical ingredient) and elution of sulfobutyl ether-[beta]-cyclodextrin by size-exclusion mechanism in one chromatographic run, was demonstrated. The strength (i.e., solvent strength) of neat tetrahydrofuran (THF) and of binary mixtures of THF and another solvent (e.g., chloroform, acetone, 2-propanol, dimethylformamide, ethanol, acetonitrile and methanol) and its effect on column characteristics relevant to size exclusion (total permeation volume, interstitial volume, pore volume and phase ratio) were qualitatively determined for a wide range of column chemistries (C18, C4, phenyl and cyano). Solvent effect on elution of PS, PMMA and NaPSS, and the effect of column temperature (within a relatively narrow range corresponding to typical chromatographic conditions, i.e., 10 °C - 60 °C) on their partition coefficients, KSEC, were also investigated. Accuracy and repeatability of average molar mass determinations (the number-average and the weight-average) for PS and PMMA polymers were demonstrated on C18, C4, phenyl and cyano columns with neat THF mobile phase, as well as with mobile phases that are binary mixtures of "good" and "poor" solvents, such THF mobile phase modified with 30% of acetonitrile or methanol. Accuracy of average molar mass determinations was better when binary mixtures were used as the mobile phase, compared to that with neat THF. SEC separation of biopolymers on diol-modified HILIC column was compared to that on the diol-coated SEC column (TSKgel UP-SW-3000). HILIC column demonstrated superiority of the linearity of the calibration plot over that on the SEC column, which exhibited a curvature. Polynomial shape of the calibration curve for the SEC column implies less inert stationary phase surface than that of the HILIC column. Common RP and HILIC columns offer an inert surface suitable for the analysis of polymers by SEC. The bonded phases show remarkable differences in size separations when binary mixtures of solvents, which differ in strength, are used as the mobile phase. The presence of a strong solvent is critical to the elimination of undesired non-SEC interactions of solutes with the column packing. The solvent strength varies, depending on the column's stationary phase chemistry. The thickness of the stationary phase determines the column's total pore volume, and changes with the mobile phase composition and with the identity and bonding density of the bonded ligands. It appears that this is related to the ability of the stationary phase to preferentially adsorb the stronger solvent and to the ability of a weaker solvent to change the conformation of the bonded ligands, e.g., from ordered state (extended with ligands aligned away from the silica substrate) to disordered, more dense state (folded, with ligands pointing towards the silica substrate). The solvent impact can be three-fold: (i) change of the polymeric coil size, and possibly shape; (ii) change of the stationary phase pore volume; and (iii) change in the solvated stationary phase properties as it may become "good" or "poor" solvent for the solutes. If the effect of this impact is properly moderated, then the greatest benefits of a wider molar mass range and an optimized solute resolution can be realized. Moreover, at a certain composition of binary mixtures of THF with other solvents, solutes of the same molar mass, but of different polarity and/or chemical composition, can be eluted at the same elution volumes, and thus allow quantitation of such solutes using a single calibration curve, as observed for PS and PMMA analytes. Temperature effect is realized in decreased elution volumes of polymers with the temperature increase. SEC partition coefficients are temperature independent in the range of 10 °C - 60 °C and therefore, elution of PS and PMMA polymers is by a near-ideal SEC on reversed-phase columns. SEC partition coefficients of water-soluble NaPSS are temperature independent in the range of 10 °C - 60 °C on the HILIC column and therefore, elution of NaPSS is by a near-ideal SEC on this column. Near-ideal SEC elution of NaPSS on cyano and C18 columns is also possible, but only at higher column temperatures (30 °C - 60 °C for cyano and 30 °C - 50 °C for C18). While elution of NaPSS on C4 and phenyl columns follows the SEC mechanism, temperature dependency of its partition coefficients implies significant non-SEC interactions even at higher column temperature (e.g., at 50 °C).

Reversed-Phase and Hydrophilic Interaction Chromatography Columns for Size-Exclusion Chromatography

Files and links (1)

Abstract

Metrics

Details

Reversed-Phase and Hydrophilic Interaction Chromatography Columns for Size-Exclusion Chromatography

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media