Liquid Chromatography Problem Solving and Troubleshooting

The first and most obvious strategy is to survey the literature to see if your analyte or mixture of analytes is similar in structure to other compounds that have been reported previously. Often, if there are not large structural differences that change the analyte’s polarity too much, then experimental conditions that have been used successfully to separate similar compounds are a good starting point. However, once you have tried this and are unsuccessful, then there are two useful HPLC approaches for establishing an initial set of starting eluent conditions. Experimentally, the quickest approach is to carry out a simple gradient-elution separation starting with a mobile phase of low eluting power and programming its strength to increase with time. A 10to 15-min linear gradient is often a good starting point. From the gradient profile it is possible to predict starting conditions from an extrapolation of the gradient profile at the time a particular solute or group of solutes elute from the column. For reversed-phase assays the starting gradient profile is run from high aqueous conditions to high organic conditions. In order of usage and increasing elute strength, the three principle organic cosolvents are methanol, acetonitrile, and tetrahydrofuran. For normal-phase separations, the gradient begins with the nonpolar carrier solvent and ends with the nonpolar carrier solvent containing a high level of a polar modifier such as 2-propanol. In selecting the polar modifier, a useful quantitative scale (i.e., the eluotropic series) is based on the work of Snyder et al. (1,2), who examined how strongly solvent molecules adsorb to polar substrates. Although this scale was developed initially for alumina, it has been modified for silica and can be found in many chromatography books. An initial mobile phase composition can also be established using an isocratic approach by starting with a solvent with strong eluting power that is systematically decreased in subsequent separations. When developing reversed-phase assays, the starting mobile phase should contain a high level of the organic cosolvent. It is advisable to begin with nearly 100% of the organic component for highly hydrophobic compounds, especially if more retentive surfaces such as ODS packings are being used. In the case of normal-phase separations, the scouting experiment starts with a mobile phase that contains a high level of a polar modifier such as 2-propanol (i.e., 20–30%) added to the nonpolar Journal of Chromatographic Science, Vol. 40, April 2002