odor sample

[ito_miura]

Dear all,

I have problem with my new sample, there is sample that have odor like ammonia, medicine, and camper. How we analyze the odor, except we use GC-MS/olfactometry ? can we analyze that with Pyrolisis-GC-MS ?
Thank you for your idea...

ito

[Peter Apps]

I cannot see any need to use pyrolisis - equilibrium headspace or purge and trap would be my first choice. If you can't do that then make a solvent extract, evaporate a couple of drops of it on a filter paper to check that it still has the odour, and then do a split-splitless injection.

Peter

[ito_miura]

Hi Peter,
Thank you for your reply!!!
We have been extract (with ethanol) that sample that have odor and the sample that doesn't have odor (for control), we analyze used autosampler GC-MS and we got the chromatogram from both sample. There are 2 different compound we met from odor sample (compared with control sample), there are 2- piperidinone, N-[4-bromo-n-buthyl]- and sulfurous acid, dodecyl 2-propyl ester (use NIST library...)
And now, we confuse how we can difine that compound are really caused that odor ...
Anybody knows how the method to knows from where both compound comes to the sample??

thank you for advance
Ito

[Peter Apps]

Hello Ito

The most straightforward way is to do the olfactometry - if the odour elutes when the peak does then it is probably causing the odour. Otherwise you will need to try to find a description of the odour on the internet, or get the pure substances and smell them (cautiously). Be aware that MS library searches do not provide definitive identifications, so you need to check at least the retention index, or better still co-inject with the pure substance to show that it co-elutes with the suspect peak.

Peter

[Don_Hilton]

One other caution: The peaks you found in the MS may or may not be the cause of the odor. I recall back in the days of the strip chart recorders, a coupleof coleagues who were doing some odor work with a GC with the efluent of the GC column split between an FID and a sniff port. Some of the many peaks in the FID trace were annotated with a description of an associated aroma. But most of the anotaions for aroma were along what appeared to be flat baseline.

[MSCHemist]

Also keep in mind that the amount of a compound is also no indication of how important it is to the smell. Some compounds you can need parts per hundred before you start to reach the odor perception threshold. Other compounds the parts per billion range are readily percieved.

I was reading about isonitriles a while back and aparently just opening the bottle in a high flow hood is enough to make eveyone in the building queesy for a week.