Chromatography Forum

Running EPA 522 (1,4-dioxane) splitless SIM-mode on an Agilent GC-MSD (5977B) nearly exactly as described in the original method.

We are having trouble with an unresolved, fairly large leading interference peak on the M+ 58 (confirmation ion) SIM that tails over the top of the smaller confirmatory ion peak at the LOQ. I am fairly sure it is a hydrocarbon of some sort. The solvents are clean, but the signal is still there for every injection. (Note that this is a peak and not a continuous background interference). I believe it is hydrocarbon-based because it also has the co-eluting M+ 43 signal.

We have tried new He cylinder, a new "large" He hydrocarbon/O2 scrubber. cleaning the injector, a new inlet liner, new septum, new column (all have been conditioned) and a new injector syringe. Also baked out the MSD and source.

Has anyone encountered this with EPA 522 or similar method? Anyone have suggestions?

Is your methanol P&T grade? Have you tried a different lot#? I know it is expensive but it is necessary.

Is your methanol P&T grade? Have you tried a different lot#? I know it is expensive but it is necessary.

Thanks for the suggestion. Yes, we have confirmed it is not the methanol or the dichloromethane used in the method. This M+ 58 peak is occurring with all injections of any GC solvents at roughly the same raw count intensity. Since it is a peak (albeit broad and tailing) and not baseline noise, I believe it is likely prior to the MSD.

How bad are we talking? I've seen a fronting peak in samples but I usually just ignore it. I don't see it in our unextracted curve and it's barely there in our blanks.

We run a really janky method based off 522 that's missing some preservatives and the surrogate became the IS, same column though.

How bad are we talking? I've seen a fronting peak in samples but I usually just ignore it. I don't see it in our unextracted curve and it's barely there in our blanks.

We run a really janky method based off 522 that's missing some preservatives and the surrogate became the IS, same column though.

This is exactly what we see! On the M+ 58 SIM (blue trace) at 7.95X min. (can't make out the last digit), the large preceding peak is unresolved and prevents the accurate integration of the confirmatory ion to determine the ion ratio as required in the method, unless we are reading the method incorrectly. We have been unable to reduce or resolve this interference peak.

Doesn't look like 522 says anything about letting analyst judgement overrule ion ratios. You could try finding a particularly bad offender and injecting a large volume in scan with an increased multiplier voltage and see what you pick up.

Are you using sodium sulfate in the extraction process?

That is notorious for having hydrocarbon impurities in it, and would be a cause for it to be in samples but not in unextracted standards.

The method suggestions of baking sodium sulfate at 400c helps but we have had some in the past that required putting it in a Soxhlet with DCM overnight to remove all the hydrocarbon interferences.

We have confirmed it isn't the sodium sulfate, but it may be present in our activated carbon used for SPE.

We have confirmed it isn't the sodium sulfate, but it may be present in our activated carbon used for SPE.

I had a lot of trouble with dirty cartridges of activated carbon when doing EPA 541 analysis for alcohols. One huge interference was 1-butanol which we found came from burning natural gas in bunsen burners nearby and also when we had a leak in the HVAC exhaust.

What are you using for the activated carbon? We're using the Waters AC-2 cartridges. We don't use any sodium sulfate in our extraction.

What are you using for the activated carbon? We're using the Waters AC-2 cartridges. We don't use any sodium sulfate in our extraction.

AC-2 were what we used for 541, but also tried ECT ones with same background. If you pull air through them they will pickup just about an organic contamination in the air, better to dry them with positive pressure UHP Nitrogen.

Perhaps a long shot. But if you have significant Argon in the carrier gas.. Might you be getting ArO formation? You would get mostly a 56 mass for O16 but you could also get a 58 mass for O18.

Perhaps a long shot. But if you have significant Argon in the carrier gas.. Might you be getting ArO formation? You would get mostly a 56 mass for O16 but you could also get a 58 mass for O18. I see this in ICP-MS but not sure there is enough energy for this to happen in GC-MS.

Hi I am experiencing a similar problem with the 58 confirmatory ion. Our peaks seem to be coeluting at low levels of 1,4-dioxane detection (<0.4 ug/L). We are also using AC-2 Waters Sep-Pak cartridges. I have seen similar issues with phenomenex activated carbon cartridges. Has anyone resolved this issue or found the source of it?