HPLC Column Selection

As chromatographers we face a bewildering number of choices when it comes to selecting the correct high performance liquid chromatography (HPLC) column for an analysis. True, we may be able to find a suggested stationary phase in the literature, but are the column dimensions applicable, will that column geometry work well with our system and allow us to attain our required detection levels? This installment of "The Essentials" establishes a number of "rules of thumb" that can be applied when selecting stationary-phase type and column dimensions to somewhat ease the burden that comes with such a proliferation of choice.

Column length is typically predicated by the resolution required from a "traditional" HPLC system and in this sense we mean systems that have not been designed or adapted to minimize extracolumn volume or cannot achieve system back pressures above 400 bar. High resolution is typically required when separating samples with many components or when many chemically similar compounds are being separated. Beware that HPLC is a diffusion-limited technique and that above analyte retention factor (k) values of around 10, increasing retention will have little or no effect on resolution because of decreases in efficiency caused by an increase in longitudinal molecular diffusion. So you may need to adjust the eluotropic strength of the eluent to reduce overall retention time.

Column internal diameter is typically chosen depending on analytical requirements and system limitations. It dictates the speed of analysis and influences method sensitivity. Use 4.6-mm i.d. columns when working with traditional HPLC systems described above, or 3-mm i.d. columns if efficiency loss due to extracolumn volume is not too great. Use 2.1-mm i.d. columns when you are using low-extracolumn-volume systems to save solvent, increase sensitivity (such as when working with a limited sample), or to reduce flow rate while still working at a reasonable eluent linear velocity (when working with electrospray ionization mass spectrometry [ESI-MS] systems, for example).

The particle size of the stationary phase support affects the efficiency of a separation and one typically needs high efficiency when trying to separate a few components in a short time or many components in a reasonable time. Smaller particle sizes bring higher efficiency; however, this is usually at the expense of an increase in system back pressure, unless flow rate is adjusted downward. The most widely used particles are 5 µm in diameter and provide very reasonable efficiencies when used with traditional HPLC systems. Other particles with diameters such as 3.0 µm or 3.5 µm provide higher resolution and shorter analysis times, but beware of increases in system back pressure and also note that as particle size reduces so does the retaining frit porosity so greater care must be taken with sample and eluent preparation. Sub-2.0-µm particles are typically used with ultrahigh-pressure liquid chromatography (UHPLC) systems, which are capable of dealing with high system back pressures (1000 bar and greater) and can generate very high efficiencies for high resolution or very fast separations. We now also have a choice in particle morphology with the availability of core–shell particles, which give rise to high efficiencies (approaching those of sub-2-µm particles) with much lower system back pressures. The lower back pressure provides the advantage of allowing their use on traditional HPLC systems.

Pore size describes the average pore diameter of the pores on the surface of the silica packing material. Pores of 6–12 nm are used with analytes below ~2000 Da and 30 nm pores are used for analytes above 4000 Da. For the 2000–4000 Da range you should experiment to see which pore size gives the best retention and efficiency. Wider-pore columns tend to provide a higher surface area and therefore are more "loadable" — that is, more analyte mass can be applied before peak shape deterioration (asymmetry) becomes a problem.

Type II (B) silicas are less acidic than Type I (A) and therefore give rise to less peak tailing when analyzing polar or ionizable analytes (especially basic molecules).

Select hydrophobic stationary phases (C18, C8) when differences in analyte hydrophobicity are large and can be exploited to affect a separation. Typically, this is when the carbon to heteroatom ratio within the analyte is large. Hydrophobic stationary phase ligands do not work well in 100% aqueous eluents and are not well suited for the retention of more polar analytes.

Modified alkyl phases (those with more polar functional groups embedded within the ligand) are used when separating analytes with different (polar) functional groups and where exploiting differences in the bulk hydrophobic properties of analytes does not produce an effective separation. Typically, these phases can be used with 100% aqueous eluent systems and are better at retaining polar analytes.

Phenyl-containing phases interact strongly with analytes containing p electron systems (aromatic, unsaturated) and will retain these analytes in preference to those with no p electrons.

Cyano and fluorinated phases interact strongly with basic, nitrogen-containing, and halogenated species and should be used when the analytes contain these species.

Amino, diol, and silica phases are traditionally used for the separation of polar analytes in both reversed-phase and normal-phase modes. Lately they have also found utility in hydrophilic interaction liquid chromatography (HILIC) mode in which highly organic eluent systems are used to retain polar analytes using polar stationary phases.

Get the full tutorial at: www.CHROMacademy.com/Essentials (free until June 20).