Help: Funny GC/MS peak and identification

[Azimpatar]

Dear all. I need your help and suggestion.

I did an experiment identifying possible compounds from marine organism and found brallobarbital (CAS # 000561-86-4) in the sample when compared to available library in MS. However, brallobarbital is synthetic drug that derived from barbiturates. It is funny to get brallobarbital from natural product like marine organism. What do you think? What should i do? Or do I mis-interpreted it?

Do the compounds are isomers that resembles brallobarbital or anything? Can you suggest anything I can explain.

FYI,the marine extract that has been prepared are an aqueous extract, dissolved in methanol prior to GCMS analysis.

Thanks you guys!

[Jetjamnong]

Hi Azimpatar,
I think the Brallobarbital is the rare synthetic drug in the market and now I saw the EI-Spectrum of this compound that show m/z 207 as base peak (when 207 is the m/z of GC column bleed too).
Do you subtract the peak background before library matching? if your spectrum is pure may be you will find Secobarbital and Hydroxyzine (the ingredient of Brallobarbital, data from Wikipedia).
The best way for confirmation is to get reference material or standard of Brallobarbital and run again (some time with -TMS)

Jetjamnong

[Don_Hilton]

1) look at the discussions on this web site about identifying compunds by mass spec library search. The greatest caution I would give here is that if you are looking for previously unidentified natural products, the spectra will not be in the libraries. But, you will get a match to the spectrum that gives the best match score - and the compund may be entirely unrelated.

2) Look to be sure that all the major ions in the library spectrum are present in the unknown. I've seen "good" match scores between complex spectra, but a major ion in the library spectrum was not present in the unknown - this allowed me to immediately reject that hit.


3) look at ion traces. Be sure they all rise and fall together. If The ions seem to have different times for the apexes or have different start or end times for the peaks, you have a mixture of compunds. In this case you need either better separation or good deconvolution software and adequate data density for the deconvolution to work.

[Jetjamnong]

Don explained very clear.

[Azimpatar]

Dear Jetjamnong & Don,

Thanks,

Azim

[ods-at-pacific]

The NIST 11 EI Mass Spectral Database has a spectrum of Cyclobarbital, which is structurally similar to Brallobarbital. The mass spectrum of both compounds is dominated by the base peak at m/z 207. There is a mass spectrum of an N,N-bis(trimethylsilyl) derivative of Cyclobarbital in NIST 11. This spectrum exhibits a molecular ion peak, and an [M – 15], and a [M – 29] peak. The [M – 29] is the base peak.

You may want to try forming a derivative of your compound. If the derivative retains the Br atom and its mass spectrum exhibits a molecular ion peak, that will be very informative.