Page 1 of 1

GC-MS of gases from fermentations

Posted: Mon Oct 05, 2009 2:32 pm
by botrytis
I was wondering if anyone has any experience with GC-MS of fermentation gases and how difficult it is. We are doing an anaerobic fermentation which will produce hydrogen, carbon dioxide, carbon monoxide, maybe hydrogen sulfide, and some volatile amines. I want to know how hard it is, what type of equipment is used, and any other information anyone can fill in.

Any information is appreciated.

Dave

Posted: Mon Oct 05, 2009 2:50 pm
by AICMM
botrytis,

Better bet would be to just use GC with appropriate detector. There are several other posts on this subject, see fixed gas analysis for example. Problems include same mass for CO as nitrogen, hydrogen being very close to helium, vacuum system constantly pulling in air....

Depending on the concentrations you are interested in, this would probably be easier with a GC and a TCD or HID detector.

Best regards.

Posted: Mon Oct 05, 2009 3:43 pm
by botrytis
The problem is - I need to find out what is in the gas besides those compounds. I know there are issues with it but I need actually figure out what the bacterium is producing. We will be asked by the EPA so this is a concern when we go to scale up this process.

So, any information and al information is appreciated.

I don't want to have to try different compounds (like standards) as we really dont have the time or themoney to check all the possible permutations.

Posted: Tue Oct 06, 2009 3:32 pm
by Ron
One thing to keep in mind is that you sometimes need more information that the mass spectrum to identify the compounds present in a mixture. There are numerous compounds with the same mass, and at times you need retention index or other information to figure out which of the possible compounds is really present. Isomers can be really difficult in this respect. As far as library searching goes, it is a good tool, but all a good match score means is that you have matched a compound in the library, whether or not that compound can possibly exist in your sample.