Troubleshooting GC/MS (single-quad) compound identification


I'm relatively new to analytical chemistry and not yet super competent in the use of the associated software. I have been spinning my wheels for the past several weeks trying to figure out how I can create a custom library for identifying the compounds present in my data output for GC/MS (single-quad) runs, and I am hoping someone can point me in the right direction for how to do this most efficiently.

I've run authentic standards for most of the key compounds I am interested in and have a spreadsheet of retention times as well as MS spectra for these compounds. I am using Masshunter Qualitative Analysis (B.06.00), which is the most recent version of software I have access to. Most of the compound identification options in the program seem to either be intended for use with a different analytical method or require a known compound mass also to match RT values, which I don't see how that information can be derived from either chromatographic peaks or EI spectra.

I tried creating a custom PCDL using PCDL Manager (B.08.00) and am able to build a custom database manually to set at least retention times. However, it seems I can't add spectra to these compounds in PCDL. When I try copying and pasting a spectrum for a compound from qual I get an error that says something about the spectrum needing to be from a soft ionization source. And then when I try in qual to identify compounds in one of my samples by retention time (actually there is only an option to identify by mass AND retention time) with my custom PCDL set as the database, none of the compounds in my list are successfully identified despite evidently matching retention times.

I figure that what I am trying to do must be possible since Masshunter can evidently be used to identify compounds using the NIST library (which I don't have access to). Is there something simple I'm missing? Or is there software out there that can do this job much more efficiently? I can annotate peaks in my chromatogram manually, but I have several dozen samples I would like to analyze and compare as quickly as possible and so an automated identification method would be extremely helpful.


  • Hello  ,

    B.06.00 is out of support at this point, but I can offer some general suggestions for current revisions and hopefully you can apply them to your version. Version 12.0 is the current release. If possible, I would investigate your options for upgrade with your local sales and support team.

    For GC/MS analysis Find by Integration or Find by Deconvolution are the primary choices for finding compounds.

    Once found, you can then perform your identification by library search. While the PCDL manager can be used to make and manage GC/MS libraries it is typically less than ideal unless you are starting with an existing PCDL library. The PCDL manager must be compatible with your version of Qual, so that may be why you are having issues using PCDL B.08 with Qual B.06. 

    For creating a custom GC/MS library with your own standards it can be easier to use Unknowns Analysis and the Library Editor. Both of these are part of Quantitative analysis. Once created, the libraries can be used with Quant, UA, or Qual. There is a general discussion in this Community post.

    (+) Best Practices - Library Searching MassHunter - Forum - Mass Spectrometry Software - Agilent Community

    I'm not certain that UA B.06 has all of the same features as current versions, but it may be worth investigating as a way to find compounds and then export them to a library.

    If you find this is not an option in your version or does not work as in current revisions, you can still manually create entries in the Library Editor using spectrum extracted in Qual. Spectrum can be copied to the clipboard in Qual and then pasted into the Library Editor after adding a new compound. At a minimum all you need is a compound name, but RT information and other information can be included. You can look at the library demo.mslibrary.xml for an example of what can be done. 

Was this helpful?