Ghost peaks related to deterioration of column?

[xileni]

Hello,
I am running some samples at a Shimadzu 2014 GC-FID (100m column) and after some samples some peaks which should be quite small appear significantly bigger, along with some ghost peaks, mainly in the end of the gradient.
Especially the last ghost peak appear with long tailing.

The last stage of the gradient I use is 30 min at 240oC, so I am wondering if high temperature for that long coulld affect the column.

The column, ferrules, liner +o-ring are new and I don;t think that the injector septa is bleeding. Also, the syringe has 3 rinses before every injection and all peaks are eluted since I have no peaks for at least 6 min in the end. The gases I use are also not conatminated.

I also thought that in the end of my sample preparation I have to evaporate the solvent A (isooctane) and redilute it in hexane as it is more friendly to the column, may be a mixed solvent could cause something like this? Though I am positive of the complete evaporation of isoctane.

The same problem appeared in the previous column which was ''older'', so when I changed it to a new one I realized that that peaks were ghost peaks.But this column have only been used for around 100 samples so far.

My concern is not to deteriorate faster the column, since I can avoid integrating the ghost peaks as long as they do not interfere with the peaks I want.

Any suggestions???
Many thanks!

[Peter Apps]

Welcome to the forum.

What kind of column is it - stationary phase, thickness ?

Strictly ghost peaks are those that elute after sitting on the column since an earlier injection.

Do you notice any change if there is a longer interval between injections ?

How do you know that your gas is clean, septum is not bleeding etc ?

Peter

[skunked_once]

I have to evaporate the solvent A (isooctane) and redilute it in hexane as it is more friendly to the column,

I have never heard of this before. Could someone please explain if this is true and if so, why?

[xileni]

Welcome to the forum.

What kind of column is it - stationary phase, thickness ?
The column is a Varian CP-Sil 88 100x0.25 (0.2)

Strictly ghost peaks are those that elute after sitting on the column since an earlier injection.

Do you notice any change if there is a longer interval between injections ?
No, interval time are still the same

How do you know that your gas is clean, septum is not bleeding etc ?
I was running some other samples just before and nothing like that appeared. Septum is relativley new <100 samples

Peter

Thanks a lot. Happy to found you.

[xileni]

I have to evaporate the solvent A (isooctane) and redilute it in hexane as it is more friendly to the column,

I have never heard of this before. Could someone please explain if this is true and if so, why?

I am not 100% sure that that is the case. I will ask the one who suggested it though and let you know

[xileni]

and the mixture of FAME standard I use it's alright, with a really small peak in the RT of the sample's ghost peak.

[Peter Apps]

So, when you inject a standard you do not see the contaminant peak that is causing problems when you inject samples ? Is this still true if you inject standard after running samples. If the standard is always clean, then the so-called "ghost' peak cannot be a ghost peak at all - it must be part of the sample.

What do you do to your samples that could cause contamination with a late eluting compound ?

Peter

[xileni]

So, when you inject a standard you do not see the contaminant peak that is causing problems when you inject samples ?no, the standard is clean of contaminants but I can see the residue of that peak only after samples with that peak. Is this still true if you inject standard after running samples. If the standard is always clean, then the so-called "ghost' peak cannot be a ghost peak at all - it must be part of the sample.


What do you do to your samples that could cause contamination with a late eluting compound ?I strongly believe that it's not a contamination through preparation, cause replicates from the same sample, in different time they appear that peaks or they dont . An assusmption of mine is that there is an overload of something that can't be eluted until the end and it appears with this patern. Maybe changing the gradient could show something? /color]

I really hope it's not degradation of the column


Peter

[Peter Apps]

OK, let's do some experiments to clear up whether the problem peaks are sample contaminants, carryover, or genuine ghosts.

In this order:

Run a sample that contains the peak that seems to be causing the problem in subsequent runs.

Run an injection of clean solvent.

Run another injection of clean solvent.

If the problem peaks appears in the solvent injections then it could be carryover or a ghost peak (it might originate as a sample contaminant, but that is a separate problem).

If the retention time in the second solvent run is the same as in the first solvent run then it is way more likely to be carryover than a ghost peak. A real ghost will not be there at all in the second solvent run.

If it is a ghost peak you can hold the column for longer at the max of the temperature programme until it elutes in the run that it was injected in.

If it is carryover you need to rinse the syringe more, and with two different solvents of different polarity, and make sure that the inlet is completely clean. Also make sure that you syringe wash solvents are clean.

If the problem peak is there in both the solvent runs, does it get smaller in the second run ? If it does it is very likely to be carryover. If not it is a contaminant in your wash solvents or your system e.g. dirty septum, dirty carrier gas.

I find that hope and belief are not very helpful in troubleshooting.

Peter

[xileni]

OK, let's do some experiments to clear up whether the problem peaks are sample contaminants, carryover, or genuine ghosts.

In this order:

Run a sample that contains the peak that seems to be causing the problem in subsequent runs.

Run an injection of clean solvent.

Run another injection of clean solvent.

If the problem peaks appears in the solvent injections then it could be carryover or a ghost peak (it might originate as a sample contaminant, but that is a separate problem).

If the retention time in the second solvent run is the same as in the first solvent run then it is way more likely to be carryover than a ghost peak. A real ghost will not be there at all in the second solvent run.

If it is a ghost peak you can hold the column for longer at the max of the temperature programme until it elutes in the run that it was injected in.

If it is carryover you need to rinse the syringe more, and with two different solvents of different polarity, and make sure that the inlet is completely clean. Also make sure that you syringe wash solvents are clean.

If the problem peak is there in both the solvent runs, does it get smaller in the second run ? If it does it is very likely to be carryover. If not it is a contaminant in your wash solvents or your system e.g. dirty septum, dirty carrier gas.

I find that hope and belief are not very helpful in troubleshooting.

Peter

Thanks a lot Peter! Unfortunately they gave me priority of running different samples for the moment, so I will do my tests and will let you know in the future about the results.
Belief is a way of explaining your sequence of scientific thoughts and hope is there when you can't explain or justify a potential result knowing that your knowledge is temporarilly insufficient