I'm trying to characterize the impurity profile of an organic liquid sample that is already described as 99.7% pure (Boiling Point 230 C).

Would a GC-MS be a viable way to identify the smaller peaks? e.g. How small a peak could one expect to get a decent mass spec. pattern for? Say, a peak with 0.01% of total peak area, would it be viable to identify it on a GC-MS?

I'm assuming the smaller peaks can be resolved satisfactorily. I don't need quantification of the peaks just identification. I can match against the NIST library of spectra but also have a clue based on the manufacturing process about what the likely impurities may be.


You can get GCMS peaks for very small components of the overall mixture.

However, most GCMS patterns are electron impact, and hence probabilistic, so if you don't have enough sample, the fingerprint won't be reliable.

In my experience 0.3% produces spectra which are 'good enough'... as long as you don't inadvertently overload your column.

| improve this answer | |
  • 1
    $\begingroup$ Of course, the background noise is important for how small peaks you can confidently identify as a "signal", and for how small peaks you can actually confidently quantify. A bunch of factors come into play that affect the resulting limit of detection and limit of quantification. $\endgroup$ – Yoda Aug 24 '16 at 19:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.