HPLC Products

Normal HPLC peaks should be sharp and symmetrical So peak tailing indicates a problem. It also makes the peaks wider, causing them to become less well resolved. And it causes the peak area to change, so quantitation becomes invalid. If you need a quick fix, change the column.

Possible causes and their solutions:

Void in the column.

If the inlet bed of the column has dissolved, the particles become smaller and smaller until they drop down between the bigger particles, leaving a void at the top of the column and restricting the flow, giving rise to an increase in back pressure. Removal of the top fitting to the column may confirm that this is the case. The symptoms of a column void are broad tailing peaks and increased back pressure.

If there is a void, the normal procedure is to replace the column. Should you be on a desert island, or working in a lab where the lack of money makes you feel like you are on a desert island, you may like to try topping up the column with spare packing material. The easiest way to do this is to remove the top end fitting from the column, thereby exposing the void. As quickly as possible, transfer some wet packing material from the bottom of an old column of the same type, leaving the silica a little proud of the end of the column. Replace the inlet fitting on the column that had the void, preferably using a new frit. This should correct the poor peak shape, although the increased back pressure will remain. For further instructions on how to do this, please call the helpline number below, or book on the HPLC Troubleshooting Course!

Silica dissolves when the pH goes over 7, at elevated temperatures, in highly aqueous mobile phases, and at high buffer concentration. If your method uses any of these, then precautions need to be taken against silica dissolution, which can include:

• Use a column which is stable at higher pH, such as Reprosil Pur Basic from Dr Maisch
• Use a polymer-based column where there is no silica to dissolve, such as Shodex ODP2-HP
• Use a pre-column before the injector to saturate the eluent with silicate anions.

Sample solvent a strong eluent

If the sample solvent is a strong eluent, in the first few seconds after injection, it drags the sample through the column, spreading out the bands, before it is diluted out of sight. This arises when difficulty is experienced getting an organic sample to dissolve in an aqueous eluent, and pure methanol or acetonitrile is selected instead. As a guide, the sample solvent should be no stronger an eluent than the eluent itself, and if using a gradient, than the gradient starting conditions.

Ion Suppression Required

If the sample is an acid or base, the dissociation equilibrium between the ionised and unionised forms gives rise to very broad and sometimes tailing peaks. The solution is to use ion suppression. This involves adding an acid or base to the eluent to force the ionisation equilibrium to the unionised form. This means that for carboxyllic acids, an acid such as phosphoric acid is added (about 10-15ml/litre), and for bases, a base such as triethylamine or diethylamine is added, in about the same concentration.

Column Inlet Bed Contaminated

If the column inlet bed is becoming contaminated with components of the sample which bind strongly to the column, the active sites at the entrance to the column become blocked. This causes the sample to spread out, passing over the first particles in the column, and giving rise to broad and usually tailing peaks.

Provided that there is no void, the solution is to invert the column (connect the inlet tubing to the outlet of the column and vice versa) and pump a strong solvent such as pure THF for maybe an hour. This will wash off most of the strongly bound contamination, thus allowing the column to function normally again. This can also be done whenever the inlet bed is seen to be discoloured. Column inversion is simply to allow the contamination to be washed off as soon as it is released by the column packing. Eluting it through the whole length of the column would take many times longer, and it would be hard to tell if it had simply been moved to the middle of the column.

If this solves the problem, you definitely need a guard column, and don't forget to change it regularly.

Various other possibilities exist, which are listed under Broad Peaks.