How to interpret the data

[zombie]

Hi,

I am a newbie in the field of GC-MS and my basics aren't clear at all. Can someone suggest me good sites from where I can pick up a few spectra and learn to interpret them? I have to interpret my data manually (not really rely on NIST library) and I am lost on how to look at the fragmentation pattern (EI), especially for poyenes......

Someone, please guide me!!!

Thanks

[Don_Hilton]

First question is what are you trying to do? Are you trying to do quantitative analysis of known compounds or qualitative analyis of unknowns?

Either way I would suggest a class in the use of mass spectroscopy. The opportunity to talk with someone who is experienced can be invaluable.

Big caution: Yes you can obtain library matches to spectra. A high match score does not confirm identification, however. There are compunds that show spectra that look similar to each other. The gamma lactones are a favorite series that I have used to demonstrate this. (The fact that you cannot rely on the NIST libraries for your work is not necessarily bad.)

If you are stuck working this out on your own, get a copy of a good book on mass spectrocosppy and work the problems in it. McLafferty and Tureček is a good book - referenced on the page: http://en.wikipedia.org/wiki/Fred_McLafferty There are many others.

For classes, there are classes ocasionally posted in the announcements portion of the forum.

[zombie]

Thank you Don. I am doing qualitative GC-MS analysis of unknowns. I agree there's nothing like talking to someone experienced, but since I can't find such people around, I have to rely on books. I have the book suggested by you, but as a beginner, many things are still hazy to me. Instead of looking at the spectra that I have obtained, I was trying to figure out the fragmentation patterns of standard compounds. Unfortunately, I am unable to understand it on my own. That's why I thought, if any of you could guide me to a site which might have lot of examples and explain how the various fragments have arisen. eg 1,3,6 octatriene gives peaks at 136, 121, 105, 93, 79, 67 and 53. 121 comes from removal of CH3 group but how does 105 peak come (loss of 16 amu)? loss of CH4? but how? some rearrangements??

Anyway, thanks a lot for your inputs.

[rp18]

Try http://www.alchemistmatt.com/#mwt
or http://www.science-and-fun.de/tools/

But you need a good book to understand basic fragmentation principles first, then go through some examples which most books have.

[AICMM]

zombie,

You can also look up spectra on the NIST web book (should Google.) So if you have compound name and/or CAS number, you can call up the spectrum from NIST and look for those ions in the chromatogram you have developed. Bit of a tedious approach but it could work for you. It would probably be worth your while to run an alkanes mix as well, especially if you are using a -1 or -5 type column. Then you can help your identification with Kovat's index.

Best regards.

[Don_Hilton]

If you are doing polyenes, interpretation is not easy. The advice AICMM gave may be a better starting place than with the mass spectrometer!

If you know that you have a polyene, use the relative retention time to estimate the molecular weight. If your compunds consist of only carbon and hydrogen, you can deduce the number of each type of atom.

On a 1 or 5 type column, more branced molecules tend to elute before less branced molecules of the same carbon number - so you can get an idea about the general shape of the molecule.

If you have McLafferty, read the first few chapters. There are one or two that describe instruments - and you can skip most of that, but it would be good to know what is going on in an EI source. Once the book gets to the chapters describing the fragmentation for specific classes of molecules, it gets tough. But you need to understand the various kinds of reactions before you can do any interpretation. So work the problems in the first few chapters. My copy is at work on my desk, so I can not look in it, but I think it is about the first eight chapters. Then, look over the chapters that cover molecules of the type you are trying to analyze.

There is a story I heard over thirty years ago - and wether it is true, I don't know. But true or false, the point it makes is a good one. The story:

According to the story, Fred McLafferty clamed that simply by careful analysis of mass spectral data, a good analyst could determine the structure of any molecule. And, it is said that his students were known to haunt the halls of Cornell at odd hours of the night, borrowing infrared instruments and NMR time - to present the good doctor with carefully rationalized spectra of varous compunds - and the rationalized spectra were then presented at various scientific meetings as proof that carefully read, mass spectral data could be used to provide complete structures.

I 1) have my doubts about the story and 2) even if true, we mere mortals tend to need some guidance in figuring out what the spectral data means.

I do not want to discourage you, but understand that spectral interpretation is a skill that takes time and effort to develop. (And, it is the kind of skill that can draw a very good paycheck!)

[zombie]

Thank you Don, rp18 and AICMM for your suggestions... I understand that good interpretation of these spectra is a skill learned over time. I will try to see if I can get something at all, but I think its hard for a non-chemist (mental-block???) !!!! And Don, you haven't discouraged me at all! What you stated is the reality...

Thanks again everyone!

[AICMM]

zombie,

Just out of curiosity, are you trying to elucidate structure or are you trying to do detection? If structure, whew, good luck. I looked up carotene on NIST and there are all sorts of bonds that want to break and the spectrum is very rich.

If detection I would offer a couple of other things to think about. Fluorescence for one, based on the structures, with good specificity. HPLC with UV-Vis detection, CI - MS to try and only get the molecular ion, NCI if you really want to get fancy. Finally, ECD will probably work really well for this class of compounds assuming you can get them to chromatograph reasonably well.

Best regards