Reducing carryover of carboxylic acids on GC/MS

[npayton]

I am working on measuring carboxylic acids (the exact type is unknown). The carboxylic acids are degradation products of a lipid. I am having carryover. I have determined the source of carryover is the syringe. I have reduced the carryover to ~6% using methanol as the solvent for both washes and by doing many (20 total) washes. I would really like the carryover to be lower, ~3%. Right now I am working with a standard mixture of acetic acid, propanoic acid and heptanoic acid. The concentration is 10ug/ml for all three components. The only other idea I have is to see if adding acid (maybe HCl?) to the first methanol wash. I dont want to inject any HCl onto my column though so HCl should not be added to the second methanol wash. The only other possibility I can come up with is to decrease the concentrations of the standards to 5ug/ml.

Does anyone have any ideas? Your suggestions would be very much appreciated!!!!

Thank you!!!

[Peter Apps]

Acidfying the methanol will make the acids less soluble, and the washing less efficient. I work with these acids over a huge concentration range and carryover has not been a significant problem, and I have been injecting solutions in a range of solvents from methanol to dichloromethane, and then washing them in the same solvents, with half a dozen pumps. Are you sure that it is the syringe that is the problem, and how do you come to that conclusion ? What kind of syringe do you have - size, plunger type etc ?. What are the wash parameters - on some autosamplers you have to set a wash volume, or fill volume. These must be at least half of the syringe volume otherwise the barrel above where the sample finally sits never gets cleaned, and the plunger is continually being contaminated by the dirty section of the barrel.

I presume this is being done by autosampler - what kind is it, what GC do you have, how are you injecting, what are the method parameters etc etc. And what else is in your samples that might be retaining the acids somewhere in the system ?

Another thing to check is that the wash solvent is actually clean, and do not top up wash vials; throw out the residue, rinse and put in fresh clean solvent.

Peter

[npayton]

Hi Peter! Thanks so much for your help. I tried 0.5M HCl in the first wash solution yesterday and it did seem to reduce the carryover. Im not sure exactly why.

To answer your questions about the specifics:

I found there was no carryover after injecting a sample if I did a blank run where no actual injection is made. There is only carryover that is apparent when the autosampler does an injection of from a solvent only vial.

I have a 10ul glass leap PAL syringe without PTFE.

I am doing ten washes from wash 1 and 10 washes from wash 2 before and after the injection of sample or standard. The full 10ul is taken up in the syringe. The instrument is a Shimadzu GCMS 2010 plus and the autosampler is a AOC-5000. I am injecting 1uL.

Parameters
50C 2 minutes
8C/minute up to 200C held for 5 minutes. I used to hold for 35 but that was not helping reduce the carryover.
Injection port 250C.
Interface Temp 230C
Ionization Source Temp 210C

I will wash out the wash vials as you suggested and try with new wash solvent. Thanks so much for all your input. If you have any further ideas that would be awesome!

Nicole

[Peter Apps]

Hi Nicole

With 0.5 M HCl in methanol you were probably esterifying the acids !

Nothing in your method jumps out as a likely cause of carryover. What kind of liner do you have in the inlet, and does it have any glass wool or other packing in it ?

Peter

[npayton]

Peter,

Sorry I meant to say the concentration was 0.05M HCl and this is only in the wash one. Wash two is pure methanol. Im not sure if what you suggested would still be likely...?

Another question- what kind of variability do you see when you inject the same sample or standard? I think I may be having too high of a variability when injecting the same standard. I thought this was due to carryover but now I am not sure. Any ideas??

Thanks

Nicole

[Peter Apps]

Hi Nicole

With 2-3 ng of short chain acids in dichloromethane we got rsds of 1-2% - this was with cold injections to a Restek Unliner in a Varian 1079, with a Restek wax column 530 um i.d. The inlet temperature was programmed at 100C/min after injection. The detector was an FID.

There is no particular reason why carryover would cause poor repeatability, unless it is so severe that it is causing a progressive increase in peak area through a series of injections - in which case you would see a definite trend in peak area.

I doubt that 0.05 M HCl would catalyse esterification to a significant extent.

What kind of liner do you have inthe inlet, and does it have any packings such as glass woll in it ?

Peter

[JI2002]

What's the solvent for the samples? Have you tried water as wash 1 solvent? What's the reason for HCL solution?

[carls]

Is the carryover level the same for all compounds? Perhaps ethanol would be a better wash solvent 1?