HS Problems

[shaun78]

I am running USP <467> and am having difficulties with the Class 2A residual solvent standard. In a nutshell, I am getting negative peaks after cyclohexane, toluene, and m/p-xylene. The negative peak comes on the tail end of the analyte peak and is making integration near impossible.

The really interesting fact is this negative peak problem just started a few days ago. I have tested water, vials, caps, and pipets as contamination sources and the results are all the same.

For what it is worth, the GC is a Clarus 500 and the HS is a HS40. Transfer line and column have been baked at 200/245, respectively, resulting in no change in chromatography.

Anyone have any ideas as to what might be causing this?

[chromatographer1]

I assume you are using an FID.

Negative peak on an FID at the end of a run could be caused by a leak.

Is the negative peak reproducible?

What happens if you run a cycle without a sample being injected?

Keep the investigation going.

Good luck,

Rod

[shaun78]

Yes, I am using a FID.

The negative peaks are reproducible but it only happens with the class 2a standard. The chromatography of all other standards are fine.

Running a cycle without the sample being injected/injecting empty headspace vial results in a relatively flat baseline with no negative peaks.

I am going to condition a new column and see if that helps out any...

Thanks,

Shaun

[chromatographer1]

the 'negative' peaks are probably the valleys between the bleed peaks which orginate from phase breakdown. All column will do it, especially silicone phases that have been exposed to oxygen and water (does that sound like a headspace injection to you? )

You could try going to a lower final temperature and allowing the late eluting peaks to elute even later to minimize these valleys.

Or you could install a new column.

The old column might perform adequately for a long time if slight modifications of the oven temperatures were made.

best wishes,

Rod

[chromatographer1]

oops, a second thought arose,

Your 'bleed peaks' might be coming from your 2A standard solution.

It might not be the column at all.

just a thought,

Rod

[shaun78]

Well, you were correct. It did happen on the newly conditioned column.

However, if this is something resulting from the pahse breakdown on the column, shouldn't I notice a similar effect in any of the other standard solutions that have been tested?

Thanks again,

Shaun

[chromatographer1]

First, Shaun, your questions are reasonable.

and second, asking people to guess causes for your problem without seeing the problem is not easy.

Sometimes it is what is IN the std that reacts with phases and can give artifactual peaks which can have negative spaces between them.

And sometimes the standard can have peaks in it that are artifacts and are not present in other standards and are not reaction products with the phase itself.

If you would make a fresh std and inject it you would probably get a better idea of the actual cause.

I hope you are able to compensate for the peaks you are seeing and are able to complete your work in due course.

best wishes,

Rod

[AICMM]

Shaun,

Are you saying a negative dip after every peak? Also, what is the concentration of the 2a standard compared to other standards? (Sorry, I am not a USP 467 expert although it seems to be a method causing a bit of grief based on the number of posts!)

Best regards.

[shaun78]

Thanks everyone for your help. I know how difficult it can be to troubleshoot problems such as this. So, again, thank you all.

For the purposes of problem resolution, it turned out that my problem was related to a dirty FID. How our FID got dirty enough to cause problems like that in about 4 months of use beats me. But it teaches me the lesson once again not to skip over the basics!

However, just bake at 350 for about an hour. If that does not fix it, then you need to replace the ceramics of the detector.

[chromatographer1]

Shaun,

Please note that the USP standard with the largest amount of chlorine gave you the problems. If a FID burns an analyte of chlorinated composition, a by-product is HCL. If there is enough silicone residue on the ceramic components enough HCl can be depositied on the 'insulators' to where they do not insulate well. I have seen strange behavior on such contaminated FIDs.

This silicone residue is usually from phase bleed or decomposed silylating reagent used in preparing samples. But since this instrument was used for headspace I suspect it is from phase breakdown of the G43 phase on the thick film column used for USP <467>. Water, oxygen (air), and temperature on a silicone phase is a receipe for phase breakdown, and have I not described a headspace injection?

I hope there was some visual indication of the detector insulator contamination. If not, then remember to schedule cleaning on a regular basis, whether you need it or not. I always prefer to keep spare parts, clean and ready, to trouble shoot inexplicable problems as you described.

I am glad you found the problem and pleased that you took the time to tell the Forum the end of the story. Thanks !

best wishes,

Rod

[shaun78]

Rod,

Thank you for the additional insight. It seems very likely that what you have described is the root problem here. I liked my old way of testing for residual solvents, which did not involve headspace but a WAX column with solvents “dissolvedâ€

[chromatographer1]

Shaun,

A 60 day cleaning of your FID might be considered more than adequate. It depends of course on your samples. You are the best judge of that. But cleaning your flame tip and replacing the ceramic isulators and polishing your electrodes can't hurt and doesn't take long to do if you have the cleaned insulators ready to install.

DMSO can foul up detectors too or any solvent which burns with a soot, like chlorinateds.

best wishes,

Rod