How to deal with system peaks - 2

[goran123]

Hi,
I am running an isocratic method, so I cannot use the trick of running a prolonged equilibration time to see if the system peaks are enlarged.
Are there any other tricks to be used in isocratic mode?
I would like to investigate if the peaks are coming from the MilliQ water used in the blank, or if it may come from the propanol or the Tris/HCl in the C18 RPC method. I will try other water qualities, but it would be nice if there would be a simple test to do ...
What about the importance of injection volumes? If I inject eg. 5, 10, 20, 40 and 80 µl of the blank, integrate the extra peak and get a nice slope and 0.99 in correlation coefficient - does that indicate that the extra peak is from the water blank, or is it because a larger injection volume would give a higher "displacement" of a contaminant (eg. from propanol) from the column?

I am grateful for all suggestions.

Regards,

Goran Karlsson
Proffice Life Science
Sweden

[DJtech]

Dear Goran,

What is the detection method?

[tom jupille]

You probably need to distinguish "system" peaks (which result from the injection upsetting the equilibrium of the system) from "garbage" peaks (which result from low-level contaminants from the mobile phase accumulating on the column during pre-equilibration). Only the latter will increase with increasing equilibration time, and they really occur only in gradient runs.

In isocratic runs, unexpected peaks can arise from one or more of the following:

- "t0 noise", generated by the equilibrium upset caused by the injection process. As the name implies, these usually occur at or near the dead time, but can occasionally be retained. With optical detectors, they typically show up as a positive/negative peak pair, but that is often masked by unretained sample components.

- "system peaks" as described above are the result of "vacancy chromatography" when an injection sample is lacking one or more components of the mobile phase. These are sometimes referred to as "water dips" because they are usually negative. If you dissolve your sample in the mobile phase and the peaks disappear, they are probably system peaks. And yes, the size *will* depend on the injected sample volume.

- "late eluters", sometimes also called "carryover" (although that can be a bit confusing) are just what the name implies: strongly retained components from a previous injection. They are positive and usually substantially wider than neighboring peaks in the chromatogram. The presence of late eluters can be confirmed by injecting a suspect sample and then extending the run time.

- "carryover", properly defined, refers to components from previous injections which are trapped or adsorbed somewhere in the system and are released during a subsequent injection. They are positive and usually comparable in width to the neighboring peaks in the chromatogram. The presence of carryover peaks can be confirmed by running a suspect sample followed by a mobile phase blank.

So, the diagnostics look like this:
- if your mystery peaks are at or near t0, you just live with it!
- if your mystery peaks are negative, they are probably system peaks. If the area tracks injection volume, and if they go away when you dissolve your sample in mobile phase, that confirms it. So long as they don't interfere with an analyte, most people just live with them.
- if your mystery peaks are positive and significantly wider than their neighbors, they are probably late-eluters. Either wait them out, improve the sample cleanup, or stick a gradient "wash-out" step at the end of the run.
- if your mystery peaks are positive and about the same width as their neighbors, the are probably either actually in your sample or diluent or are carryover from previous samples -- in which case check things like the autosampler needle rinse.

[goran123]

Dear Goran,

What is the detection method?

Absorbance at 215 nm, so most components will give a response at that wavelength.

Regards,
Goran

[goran123]

You probably need to distinguish "system" peaks (which result from the injection upsetting the equilibrium of the system) from "garbage" peaks (which result from low-level contaminants from the mobile phase accumulating on the column during pre-equilibration). Only the latter will increase with increasing equilibration time, and they really occur only in gradient runs.

In isocratic runs, unexpected peaks can arise from one or more of the following:

- "t0 noise", generated by the equilibrium upset caused by the injection process. As the name implies, these usually occur at or near the dead time, but can occasionally be retained. With optical detectors, they typically show up as a positive/negative peak pair, but that is often masked by unretained sample components.

- "system peaks" as described above are the result of "vacancy chromatography" when an injection sample is lacking one or more components of the mobile phase. These are sometimes referred to as "water dips" because they are usually negative. If you dissolve your sample in the mobile phase and the peaks disappear, they are probably system peaks. And yes, the size *will* depend on the injected sample volume.

- "late eluters", sometimes also called "carryover" (although that can be a bit confusing) are just what the name implies: strongly retained components from a previous injection. They are positive and usually substantially wider than neighboring peaks in the chromatogram. The presence of late eluters can be confirmed by injecting a suspect sample and then extending the run time.

- "carryover", properly defined, refers to components from previous injections which are trapped or adsorbed somewhere in the system and are released during a subsequent injection. They are positive and usually comparable in width to the neighboring peaks in the chromatogram. The presence of carryover peaks can be confirmed by running a suspect sample followed by a mobile phase blank.

So, the diagnostics look like this:
- if your mystery peaks are at or near t0, you just live with it!
- if your mystery peaks are negative, they are probably system peaks. If the area tracks injection volume, and if they go away when you dissolve your sample in mobile phase, that confirms it. So long as they don't interfere with an analyte, most people just live with them.
- if your mystery peaks are positive and significantly wider than their neighbors, they are probably late-eluters. Either wait them out, improve the sample cleanup, or stick a gradient "wash-out" step at the end of the run.
- if your mystery peaks are positive and about the same width as their neighbors, the are probably either actually in your sample or diluent or are carryover from previous samples -- in which case check things like the autosampler needle rinse.

Hi Tom,
Thank you very much for all information and advice. It was very useful.
Regards,
Goran