8270 Internal Standard Unstable

[sw340496]

I am having a problem with my internal standard recoveries not being stable and varying widely throughout the day. The internal standards always start with a very high recovery in the 1st run of the day and then decreasing in each successive run before eventually leveling off later in the day. This is only affecting the internal standard compounds, the target compounds are not affected and are very consistent. It looks like this problem is only affecting the deturated internal standard compounds. We are using Restek Cat# 31886 as our internal standard at a concentration of 40ppm and we are running a pretty full 8270/Appendix IX list with about 100 targets. We are running an 6890 GC with CTC A200S autosampler and a 5973MSD. We are using Restek sky precision split liners and also have the easy twist off inlet installed however I don’t think these may be of any consequence.

[rb6banjo]

So the dioxane is showing the same behavior as the PAH's? What happens if you add something like n-decane or n-dodecane to your mix and inject that over the course of a day?

[sw340496]

Yes, I am seeing the same behavior across all of my compounds. If I re-inject my standard multiple times only the internal standards will have a significant change, most of the targets are almost identical in response area throughout the day. my I'll have to see if I can find a hydrocarbon standard and see if that tells me anything.

[rb6banjo]

Odd. I thought maybe the PAH's were showing some adsorption that would not be evident for the dioxane. Stupid question but is the internal standard added to the sample in the vial? If yes, is the vial well mixed before carrying out the analysis? We had a thing happen in our lab not long ago where an intern was analyzing some vials of samples that had been frozen previously. The person did not thoroughly mix the samples before putting them on the autosampler and the results were very erratic because they stratified during freezing.

[sw340496]

Yes, the IS is added directly to the vial from the stock bottle with no intermediate dilution. I do believe that vials are well mixed. It appears that the problem is somewhat realated to the GC/MS sitting idle for any length of time (matter of hours, not days) because when we try to analyze samples/standards after the GC sitting idle we have to start the process again of doing injections and waiting for the IS area counts to come down again. We have even used the same vial as the previous day to rule out anything that is sample related.

[Peter Apps]

"this problem is only affecting the deturated internal standard compounds"

makes me think that the MS tune is drifting - deuterated ISs differ only in mass compared to anything else. If you were getting exchange of the deuterium with other components then you would see low recoveries when you re-used the previous day's vial. What happens if you re-tune in the middle of the day ?

What do you mean by "recovery" - are you comparing the size of the IS peaks with size of other peaks ?, if so is it possible that the other peaks are getting bigger rather than the IS peaks getting smaller ? Peaks getting bigger as active sites fill up is quite common, but not peaks getting smaller.

Peter

[sw340496]

When I am talking about "recovery" I am referring to area of the IS peaks when compared to the IS peaks observed in the calibration curve. The other targets really are not changing in size much if it all when observed by overlaying the chromatograms on to of each other. For example injection #1 of a sequence will have a area count "recovery" of 200% of what was observed in the curve whereas injection #5 of that same standard would be 90% when compared to the curve and then would stay stable at that area count. The other compounds would stay at a near constant area count as the IS areas are changing. When leaving the GC idle for a few hours (overnight) would require this process to be repeated.

[Peter Apps]

So the deuterated IS peaks halve in size during the day - a pretty marked change !

Is there anything in the mix that is an undeuterated analalog of an IS ? Is there is, and its area is constant then this has to be related to molecular mass, and that must mean that the problem is in the MS, not the GC.

Peter

[Bigbear]

Has the system performed correctly ever?
If yes, I would look at the injection port. I run 525.2 and when I have similar problems I "gun brush" the inlet weldment, instructions on Agilent's web site. Correction what I really do is swap my weldment out for a cleaned one. Then I clean the dirty one with the alumina slurt and Q tip.
When doing injector maintenance do not over look the split vent line (Cu 1/8" tube ) this can be cleaned, but better to swap it with new GC grade tubing cut to the approximate length of the old.