FAME interference

[DaBrainz]

I am hoping somebody here with a lot of FAME experience can help me.

I am missing my deadline trying to get this method developed for release testing for a freeze dried API of mycobacteria tuberculosis in phosphate buffer.

My method involves drying 0.5 mL of sample (equal to about 1 mg total protein content) adding 1 mL 14% BF3 in methanol and incubating for 20 minutes. The FAMEs are subsequetly extracted in 1 mL heptane and 1 mL 5% NaCl. All of this is done in an 11 mL teflon screw cap tube. The heptane is tranferred to an auto sampler vial.

The GC column is a agilent DB-225 30m, 0.25 mm, 0.25 um. I am running 2 uL splitless injection with purge at 1 min with FID detection.

The main fatty acids I'm quantifying are palmitic, stearic, Oleic and tuberculostrearic.

Now for the problem.
I am getting small peaks in my blank at the retention times of palmitic and stearic. The blank is the sample prep done in an empty tube. However I don't see any peaks if I just inject heptane. I have tried different solvent extractions such as hexane, cyclohexane and chloroform. I have tried 1% sulfuric acid instead of BF3. I have tried new columns and the peaks are there on the first injection.

The peaks seem to get smaller if I use water in the extraction instead of water.

Does anybody know why these peaks are in my matrix blank but not the neat-heptane? I've been killing myself trying to get this method out of my hair and into QC.

[chromatographer1]

The peaks seem to be coming from your reagents or glassware.

Single elimination and substitution will find the source.

Your one statement appears to be a typo. Would you clarify?

"The peaks seem to get smaller if I use water in the extraction instead of water. "

best wishes,

Rod

[DaBrainz]

I meant the peaks are bigger if I use 5% NaCl in the extraction instead water. To give you an idea of the magnitude, the range of my method is 1-100 mcg/mL and the interfering peaks are from 0.1 to 0.3 mcg/mL.

I have changed every reagent possible. The only glassware I use for the spl prep are the teflon lined screw cap vials. They are new and I only use them once. I tried sonicating them with heptane and other solvents to see if it would remove any leachables but it did not help.

[Consumer Products Guy]

Do you think you have soap (fatty acid salts) contamination in in your inlet that is being esterified in-situ by reagent blank, but not by straight heptane?

[DaBrainz]

I considered that. I have rinsed the inlet and liner with solvents and did this with different types of liners as well. I have also tried different brands of inlet septum. Not to mention its consistent from the same solution in the same autosampler vial from day to day. This also occurs on 2 different instruments (agilent 5890 and 6890).

Is it possible that the glass tubes I'm using can still leach these contaminates after a solvent rinse with sonication? The tubes are from wheaton.

[chromatographer1]

Fatty acid salts would not be rinsed from glassware by using heptane. Try using a wash of acetic acid glacial instead, or 6N sulfuric acid solution.

best wishes,

Rod

[Don_Hilton]

Consider all the materials that come into contact with the samples or reagents. I have some analytical work where fatty acids and some kind of steriodal stuff shows up in blanks, but not in solvents taken right out of the bottle. I have assumed that this is from residues left from manufacturing or packaging the lab supplies - and I have no idea which ones.

[Peter Apps]

The most likely source for these two acids is fingerprints, but they could be getting onto sample contact surfaces at any stgae of the manufacturing, packing and shipping process, so you need to clean everything immediately before it is used.

For the tubes try a series of solvents - e.g. methanol, acetone, heptane, acetone, methanol, or precede that with a lab detergent in hot distilled water. Rinse very thoroughly with distilled (or water purifier) water from a vessel that is itself clean, then dry at the highest temperature that the tubes can stand, preferably 450C or above since this will oxidise anything organic. You also need to clean the caps - same washing but a more gentle drying. When I say that you have to do this I mean that you have to do it - do not delegate it until the troubleshooting is done.

Then discard all the rinse solvents on your autoinjector, clean the vials (do not just take new ones, how do you know that they are clean ?) and refill with clean solvent. Increase the syringe rinse cycles and use two complementary solvents. Clean the needle guide.

How are you measuring and dispensing solvents, reagents and samples - is all that hardware clean ?

Try specially cleaning some autosampler vials, then use them without caps - clean blanks will show that the vials are the source of the problem.

Peter

[DaBrainz]

Just thought I would give you guys an update on this method. I tried all of your suggestions and I appreciate your help. I was able to reduce the interfering peaks to about 0.05 to 0.15 mcg/mL. By incubating the tubes at 100 degrees with concentated sulfuric acid for 30 min. The interfering peaks are now consistant enough that I have put controls in the procedure to have any interfering peaks less than 0.2 mcg/mL as part of system suitability criteria and our QC/QA people have agreed to go along with that approch.

It still boggles me why acid washing would reduce the interfering peak size but multiple acid washes will not eliminate it completly. Anyways the method is being validated this week and the results look great so far so I'm ready to put this project in my rearview.

[Peter Apps]

Thanks for the update. The failure of multiple washings of the tubes to further reduce the contaminants probably means that you have the same contaminants on something else.

Peter