Contamination?

[redpanda]

I have been running samples over the last month and my blanks are looking worse and worse. I have posted the chromatograms below. In the beginning the chromatogram looked fine, I was only running FAME standards. We started doing a few derivatization with olive oil and finally with extracted liver lipid samples. This was the acidified methanol protocol followed by extraction with hexane. The blanks have been getting progressively worse! My experience with GC/MS involved having someone looking over my shoulder the entire time I was doing something, now I am all on my own and I am lost! Any help would be great.

Everything is in hexane.

Column: HP-5MS 30 m x 0.25 mm x 0.25 μm
Injection: 1 ul
Method: 50oC 1 min, 25oC to 175oC, 4oC/min 235oC
Split: 1/50
Carrier: He


Blank before the first FAME run


Blank after the first FAME run


Blank after running about 30 samples.

Thanks

[Peter Apps]

Have you put in a clean inlet liner ?

Peter

[Steve Reimer]

Acidified methanol? If you get any of the acid on your column it will damage it and the result is a rising baseline with temperature and increasing noise. The noise is your column phase. Start with Peter's suggestion. If that doesn't fix it clip the first meter or so off the column. It is possible that your column is history.
Hexane shouldn't carry along the methanol/acid from the derivatization but the consequences of injecting acid are dire.

[redpanda]

Thank you for the replies. I will do some maintenance and replace the liner, septum ect.

Is there anything I can do to prevent build-up of this junk prior to injecting my sample. When I did LC/MS I would use a C18 cartridge desalting column on my LC prior to the sample going onto my analytical column. Is there anything similar that I can use on the GC? I will be running a lot of lipid samples from different mouse tissues and I would like to avoid this problem in the future.

[brewerbill]

To reduce the amount of acidic methanol on the column, add water to the cooled reaction mixture before hexane extraction. If the reaction is room temperature when the water is added your FAMEs won't hydrolyze to free fatty acids, and the presence of the water will reduce both methanol and titratable acidity in the hexane layer significantly.

However, this looks to me like a dirty liner. Depending on how much cleanup your samples require you may conclude that the best option is simply to replace it more often. One cleanup possibility is to load the sample onto a silica SPE and then elute the FAMEs with 9:1 hexane:diethyl ether while polar impurities like cholesterol stay on the SPE cartridge. If you go this route, quantify your FAMEs using an internal standard like methyl tricosanoate added prior to SPE cleanup.

[James_Ball]

Are you using H2SO4 or HCL as the acid in your acidified Methanol?

We found that using HCL will give less interference on the ECD than H2SO4 and it will not breakdown the column as much also if there are traced of it left over.

[redpanda]

I have been using H2SO4. I don't think the acid was the problem. I replaced the septum, inlet filter, and o-ring. The o-ring and septum were filthy and I don't know if they have been replaced since the instrument was set up about 3 years ago. It's been a paper weight until the last 6 months.

I also don't know how much of a difference this makes but the inlet filter was suppose to be for a splitless injector even though it's been running in split mode.

After running 5 blanks the chromatogram is almost back to baseline.

[Peter Apps]

I have been using H2SO4. I don't think the acid was the problem. I replaced the septum, inlet filter, and o-ring. The o-ring and septum were filthy and I don't know if they have been replaced since the instrument was set up about 3 years ago. It's been a paper weight until the last 6 months.

I also don't know how much of a difference this makes but the inlet filter was suppose to be for a splitless injector even though it's been running in split mode.

After running 5 blanks the chromatogram is almost back to baseline.

Good result ! Thanks for the feedback.

Peter