Chromatography Forum

I've been running distillation extracts on a GC-MS system using the Adams Library method to help identify volatile compounds both by retention times and spectra. I had no issues until one day when I realized that trans-sabinene hydrate, while there as a peak at the correct retention time, it was coming up as y-terpinene in the libraries.

I ran standards for both, did an overlay, and confirmed that the sample had a peak at where the standard for trans-sabinene hydrate came out, but when I checked the spectra for the standard, it too had changed to look more like y-terpinene.

Basically without change to tune, gas, or literally anything that I'm aware of, during a sequence of samples running overnight the main identifying 43 m/z of trans-sabinene hydrate dropped down to a point where it was no longer identifiable by spectrum.

Here's a picture with the two spectra of that compound for comparison. Top was from the standard I ran early on, bottom is what I started getting and haven't been able to resolve:

https://imgur.com/qIXTtLm

There was a drop in sensitivity when this happened so I figured dirty ion source, cleaned it, got it all back up and running with autotune, seemed good and re-ran the standard but nope. Aside from the sensitivity issue being resolved, the spectra looks just as bad as it did before.

Any ideas as to what it could be? I'm assuming it's not anything on the GC side since the retention time for peaks in the chromatogram are unchanged.

Also to note, the standard for 1,8-cineole looks just fine, and some other standards like nerolidol do not look fine, so it's not necessarily a specifically 43 m/z issue, nor is it a single compound issue. It's just that the primary compound I'm using as an example was the first time I noticed the change and the easiest one to test quickly since it elutes within 10 min.

I don't have an answer as to why this change happened, but it might be of interest that your instrument seems to not be the first to produce a spectra for that compound that looks more like the unexpected spectra (at least to my eye) if I have the CAS# for the chemical correct -

https://webbook.nist.gov/cgi/cbook.cgi? ... #Mass-Spec

hopefully more insights to follow as the change itself is interesting. Perhaps 43 being a common ion amongst potential contaminants is involved?

That is interesting. We have an older version of NIST installed and for whatever reason it still pulls up as y-terpinene and a bunch of other compounds. When I work through the software library backwards, as in looking for trans-sabinene hydrate first, their version of the spectra is identical to the one you linked, so I'm not sure of why the software wasn't pulling that back as an option.

Adams doesn't return either cis-/trans- sabinene hydrate but the retention time fits and hasn't changed since "the change" in the spectrum occurred.

Issue has basically been resolved! It was a combination of re-occurring air leaks and bad filaments. Resolving the air leak and installing a new filament has seen marked improvement. Thanks everyone who helped me brainstorm all this and relaying their own experiences.