J.J. Kirkland

Optimization of solvent strength and selectivity for reversed-phase liquid chromatography using an interactive mixture-design statistical technique

Abstract A general scheme combines the Snyder solvent selectivity-triangle concept with a mixture-design statistical technique to optimize the strength and selectivity of mobile phase solvents for reversed-phase liquid chromatography (LC) separations. In particular, a new method of data analysis called overlapping resolution mapping (ORM) shows advantages over previous chromatographic optimization schemes. The approach can be used to achieve a minimum resolution of all components of a mixture or, alternatively, a single pair or several different pairs of compounds within the mixture. In reversed-phase separations of nine naphthalene compounds substituted with different functional groups, tests with mixtures of methanol—, acetonitrile- and tetrahydrofuran—water that have significant selectivity diff erences revealed that no single organic modifier in water could separate all components. However, when data from seven separation experiments were analyzed with the interactive computer routine, an optimum solvent mixture predicted that subsequently gave complete isocratic separation of all components. While anticipated selectivity changes were found with aqueous mixtures of single organic solvents, aqueous binary or ternary mixtures of these organic solvents exhibited amonalous behavior toward certain solutes. In addition, individual solvent strengths were sometimes different from those predicted by previous studies. Tests on literature LC data set using simulated solvent mixtures with fifteen compounds, some exhibiting peak crossovers with diferent solvent mobile phases, clearly demonstrated the advantages of the mixture-design ORM method over other chromatographic optimization techniques.

Comprehensive characterization of some silica-based stationary phase for high-performance liquid chromatography

Abstract Comprehensive studies have been carried out on the effect of the silica support in preparing optimized bond-phase packing for high-performance liquid chromatography. Different covalently bonded reversed-phase silica-based chromatographic materials have been investigated with respect to lingand density, residual silanol group type, and concentration. Results from diffuse-reflectance Fourier-transform infrared spectroscopy, elemental and thermogravimetric analysis, and1H solid-state- and29Si cross-polarization magic-angle spinning nuclear magnetic resonance techniques were combined with liquid chromatographic data to provide quantative insights regarding the properties of silica-based bonded-phase packings. We have determined that unfavorable adsorption of basic compouds and low hysdrlytic stability of alkyl bonded-phase ligands can attrilbuted to the existence of siolated, non-hydrogen-bridged, high acidic SiOH groups. Contrary to comon belief, a stable silica support with the lowest adsorptivity for basic compounds must contain the highest possible (and not the lowest) concentration of homogeneously distributed, associated or bonded SiOH groups to ensure a minium concentration of deleterious highly acidic, isolated silanols. Under these conditions, silica-based bonded-phase packings exhibit higher hydrolytic staility and a substantially lower order of adsorptivity towards basic organic compounds.

Effect of column degradation on the reversed-phase high-performance liquid chromatographic separation of peptides and proteins

Many reversed-phase separations of proteins and peptides are currently performed in acidic mobile phases, e.g., 0.1% trifluoroacteic acid in water (pH 2) with organic modifiers. Such conditions are known to promote the cleavage of the silane from the silica in bonded-phase columns, especially for monomeric stationary phases. The stability of some columns commonly used for proteins and peptides has been examined, and it has been shown by both chromatographic and elemental analysis that degradation occurs very rapidly with fresh, totally covered column materials. Despite the loss of over half of the bonded phase in some cases, certain columns still exhibit adequate chromatographic performance, although reproducibility can be affected. The implications of these results with respect to both bonded-phase synthesis and mechanistic interpretation of chromatographic data is discussed.

Simultaneous selectivity optimization of mobile and stationary phases in reversed-phased liquid chromatography for isocratic separations of phenylthiohydantoin amino acid derivatives☆

A mixture-design statistical technique has been used to optimize simultaneously the selectivity of both mobile and the stationary phases for the isocratic high-performance liquid chromatographic separation of phenylthiohydantoin derivatives of the 20 common amino acids. This approach permits the fine tuning of selectivity to achieve the rapid separation of this relatively complex mixture with maximum resolution between the various components. An optimum isocratic reversed-phase separation has been achieved in 13 min with a minimum resolution of 1.1 for all component peak pairs. The system uses an optimum combination of four mobile phase solvents (aqueous acids, methanol, acetonitrile and tetrahydrofuran) and three stationary-phase packings (C8- CN-, phenethyl-modified silica) to obtain various separation goals.

Practical optimization of solvent selectivity in liquid-solid chromatography using a mixture-design statistical technique

A systematic approach is described for the optimization of solvent selectivity in liquid-solid chromatography (LSC), with emphasis on changes in selectivity as a result of variation of mobile phase composition. Major contributions to selectivity are provided by solvent-solute localization and solvent-specific localization. Exploitation of these effects is achieved by the use of a mixture-design statistical technique to minimize the number of experiments to find an optimum solvent mixture for LSC separation. Quaternary-solvent mobile phases are required for difficult separations to invoke the full range of selectivity effects possible for LSC separation. The four preferred solvents for LSC optimization based on localization effects are hexane, methylene chloride, methyl tert.-butyl ether and acetonitrile. In the optimization process retention data are required for only seven mobile-phase systems, and an overlapping resolution mapping (ORM) technique of data analysis is used to establish the optimum solvent mixture for the highest resolution of all adjacent bands in the chromatogram.

Theoretical basis for systematic optimization of mobile phase selectivity in liquid-solid chromatography : Solvent-solute localization effects

Abstract The optimization of retention in liquid-solid chromatography (LSC) is explored in the present paper. Previously it was shown possible to calculate solvent strength (e0 values) for multi-component mobile phases, and specifically for quaternary solvent mixtures Aue5f8Bue5f8Cue5f8D. With e0 held optimum and constant for a particular sample, the composition of Aue5f8Bue5f8Cue5f8D can be further varied for optimization of separation factors α (solvent selectivity) for various solute-pairs in the sample of interest. The selection of optimum pure solvents Aue5f8D for this purpose and the systematic variation in the proportions of these solvents for optimum separation are approached here in terms of a fundamental description of how solvent selectivity arises in LSC. In this paper we discuss two major contributions to solvent selectivity: solvent/solute localization and solvent-specific localization. In a later paper we apply these findings for the development of a systematic approach to the optimization of retention in LSC separation.

Porous silica microspheres for high-performance size exclusion chromatography

Abstract Rapid and precise high-performance gel permeation chromatography of polymers has been accomplished with columns of 6–9-μm trimethylsilyl-modified porous silica microspheres with narrow pore-size distributions over the range 60–3500 A. A 60-cm, five-column combination of this pore-size range exhibited a plate count of 2.5-104 for a totally permeating solute (toluene) and 104 plates for a 105 molecular weight polystyrene. This column set has an overall fractionation range of 102->7·106 for polystyrene, and its high resolution permits the determination of molecular weights in a 15-min analysis with an accuracy of about ± 2%, without peak dispersion corrections. The high resolving power of the microspheres, coupled with excellent mechanical strength and high purity, provides superior systems for rapidly characterizing both high-molecular-weight polymers and mixtures of small molecules by the size exclusion method.

Optimization of mobile phases for multisolvent gradient elution liquid chromatography

Abstract Optimization procedures which have been previously developed for isocratic separations in liquid chromatography (LC) have been extended to include gradient elution systems. A multisolvent classification system is used which defines LC solvent systems based upon both solvent strength and selectivity considerations. A systematic experimental design is employed to gather basic retention data on the compounds in a mixture of interest. The data can then be fitted to a second-order polynominal surface and an overlapping resolution mapping technique is used to predict the optimum solvent system for selectivity purposes. Optimization of isoselective multisolvent gradient elution systems is the easiest and should be the most useful technique. A more powerful, but somewhat more complex, selective multisolvent gradient elution system is also described.

Porous silica microsphere column packings for high-speed liquid—liquid chromatography

A new column packing for high-performance liquid chromatography, porous microspheres of silica produced by the agglutination of colloidal silica particles, has recently been introduced for use in adsorption chromatography. The narrow-size range, relatively homogeneous pore structure and short diffusion path lengths of these <10-μ particles result in very high column efficiencies, and the relatively large, highly available surface area provides for high sample capacity. The microsphere packing displays retention and efficiency characteristics which are less dependent on water content than wide-pore silica gel. Columns of the microspheres may be prepared which are reproducible in chromatographic performance, using a simple high-pressure slurry-packing procedure. More than 10,000 theoretical plates have been obtained on a single 25-cm-long column of 5-μ microspheres at carrier velocities of about 0.7 cm/sec. Plate heights of about five particle diameters and more than thirty-six effective plates/sec have been demonstrated for solutes with capacity factors (k′) in the 2–5 range. These columns may be connected in series using low-volume fittings with little loss in efficiency. Columns of the 5-μ particles appear to be limited by mobile phase mass transfer effects, contrasted to the stagnant mobile phase mass transfer limitations exhibited by similar 8- to 9-μ particles.

Method development in high-performance liquid chromatography using retention mapping and experimental design techniques

Abstract Method development in high-performance liquid chromatography (HPLC) using retention mapping and experimental design techniques is reviewed. The general strategy of overlapping resolution mapping is overviewed. A summary of various applications is examined for reversed-phase, normal-phase, ion-pair, and gradient elution HPLC, as well as stationary phase selectivity. In addition, numerical criteria for separation and optimization are detailed and a discussion of peak tracking and software is included.