You developed a nice gradient method for your pharmaceutical
impurities analysis. The peaks are sharp and symmetrical. The separation is
good. The only problem is the baseline, which slopes and makes accurate
quantitation of your peaks difficult. There are a couple of ways you can avoid
this problem and you might find one easier than the other depending on your
situation.
The
first way is to run a blank. Typically this will be your sample diluent with nothing
else in it. You only need to do this once for the method and then you can use
it for each sample. Most modern chromatography software has a feature called
“blank subtraction.” Here, you would load your sample chromatogram with the
sloping baseline and then apply the blank subtraction operation. Every
chromatographic feature in the blank will be subtracted from your sample data,
resulting in a completely flat baseline. Some analysts may find this cumbersome
though because it requires you to run an additional sample. Still, many protocols
may require the analyst to run a blank anyway to demonstrate that they don’t
have any “ghost peaks” in their sample chromatograms.
The
second method doesn’t require an additional run but it can be more tricky to initially
develop. This technique is called “absorbance matching.” The reason the
gradient slope occurs is because of a difference in UV absorbance between the A
and B solvents. You don’t notice it in isocratic methods because the solvent
composition never changes. Generally, the B solvent will absorb more because of
the organic solvent (e.g. acetonitrile, methanol, etc.). In this approach, the
goal is to achieve identical absorbance between the two solvents. The two don’t
have to have the same absorbance across the whole UV spectrum, just the
wavelength you are operating at. To achieve this, add a UV-absorbing species to
the A solvent until you don’t see any change in the baseline. Once you know the
correct amount, the process should be easy to write into your SOP. Generally
though, determining the proper amount is accomplished by trial and error. The additive
should be unretained and should not interact with or affect the sample.
Examples include nitrate, nitrite, azide compounds, etc.
An example of blank subtraction
