Reduced ISTD peak on 1 GCMS but normal on another

[pnwquake]

I have a GCMS method that has been validated and in use for many years. Recently, one of 4 internal standards has intermittently had a reduced peak size (sometimes 50% of what is expected) for samples run on a certain GCMS instrument (the same exact sample will have a normal peak size on a different GCMS instrument of the same model, which rules out any problems due to the sample itself). The instrument methods are exactly the same (I even re-loaded the other instrument's method just to be sure - it had no effect on the problem). The MS source was recently cleaned, both filaments are new, the column is brand new (I thought maybe a new column would fix the problem but it did not) and the tune is near perfect. The vacuum is good, the air and water check good. All of the other peaks in the sample appear normally. What am I missing? All I can think to do is re-clean the MS. Any suggestions would be appreciated, thank you!

[Don_Hilton]

Inlet conditions?

Is the compound with the peak that changes acidic or basic? The Grob test mix tests for inlet or column active sites with a nitrophenol and a nitroaniline. Preferential loss of one compound indicates activity.

Depending on the sample queue, would it be possible that residues would be left in the inlet that could provide activity? Are you using the same liner in each instrument?

Hot spots and cold spots related to a transfer line can be an issue for high boilers. Is the compound that varies a late eluter?

Liquid injections from a syringe or are we looking at a gas sample from a sampling loop?

If you can tell us what the internal standard compounds are and the sample matrix, it might help get a better focused answer. And GC/MS conditions could help as well - including make and model of instrumentation.

[James_Ball]

I agree with what Don said, also have you tried swapping the inlet liners between instruments to see if the problem moves to the other instrument?

Also, if it is an Agilent instrument, have you cleaned the split line? It is the copper line connected to the side of the injection port. I have seen times when one of those causes breakdown, even though it seems to be down stream of the inlet. Just take it off and flush it with solvent until it runs clean, and use a swab to clean the connection it attaches to on the inlet.

Another thing to try if the inlet itself could be clean is cleaning it with a bronze bore brush, like the ones used to clean rifle barrels. Agilent require one for a 32 caliber(0.32 inch). We also often use a brush or folded pipe cleaner attached to a Dremel rotary tool with polishing compound. This will mirror polish the inside of the injection port but does require removing the port from the instrument so you can sonicate it to remove all of the compound.

[Yama001]

James, what is the polishing compund you use? I see Wenol used in source cleaning by some enthusiasts.

[James_Ball]

James, what is the polishing compund you use? I see Wenol used in source cleaning by some enthusiasts.

I have used alumina, the type supplied with the old Agilent MS units at install, mixed with Citranox, also jeweler's rouge but you have to remove the wax base. One analyst here has even used Mother's Mag and Aluminum wheel polish with success. The only problem with the automotive types is they are in a oily base and it is more difficult to clean all the residue off.

I mostly use the alumina and citranox myself for cleaning just about everything in the injection port or the MS source. Cleans up easy by rinsing with tap water then sonicating a few times with DI water then with Methanol. Other solvent if they are compatible with your analysis(not good to use Methylene chloride if using an ECD, ect)