Problem with resolution

[michael raharja gani]

Is the resolution peak in chromatogram from analysisi crude extract using hplc must has value 1,5 or greater?

[tom jupille]

Resolution of 1.5 is often referred to as "baseline" resolution, but that term is somewhat misleading. If you have two equal-size, Gaussian peaks with resolution = 1.5, there will be approximately a 1% overlap between the peaks. Therefore, Rs = 1.5 was originally called "99% baseline resolution". With time, the "99%" part dropped off and people started referring to simply "baseline resolution".

So there is no single answer to your question. If you are collecting fractions and purifying a crude extract, you may be able to get away with less than 1.5 (especially if you will be doing further processing). On the other hand, if you are looking at trace components, then you may need much higher resolution in order to get acceptable accuracy (e.g., Rs of 4, or 5, or . . . ).

[michael raharja gani]

If i want to determine the concentration from marker of the crude extract, what is the better resolution for the marker peak?

[tom jupille]

I'll repeat what I said before:

So there is no single answer to your question. If you are collecting fractions and purifying a crude extract, you may be able to get away with less than 1.5 (especially if you will be doing further processing). On the other hand, if you are looking at trace components, then you may need much higher resolution in order to get acceptable accuracy (e.g., Rs of 4, or 5, or . . . ).

So, how accurate/precise do you need to be? What is the relative size of the closest interfering peak?

If you can tolerate errors of 10% or more and the closest interference is comparable in size or smaller than your marker, then you might be able to get away with Rs of 1 or so. If the marker is much smaller than the closest interfering peak (say, a 1:10 ratio for example) then Rs of 2 would be a good target. If the marker is very small compared to the interfering peak (say, 100:1), then a resolution of 4 or more may be necessary. All of the above are crude estimates and assume that your peaks are well-formed and symmetrical. If the peak shape is non-ideal, then higher resolution will be required.

[michael raharja gani]

Oh i see,thank you