Column for O2 analysis? And Q on GC use.

[jearles]

I'm trying to help someone else do some analysis on some samples and could use some help asap! (cuz lets wait til the week we want to run to bring this up! argh). She's wanting to analyze just O2, I believe from some soil samples she's incubating in serum vials (so likely to have other gases, such as N2, CO2, etc). Any suggestions for a column to seperate O2 presumably from N2, and in a fairly short run? I have a column she could borrow but its like a 20 min run which is unecessary for her use (its a haysep db), and she has so many samples it wouldnt' be realistic in given time frame (see below). One thought I did have is that I have an older one that's a spare, would it be possible to cut it down to create a shorter column and thus shorter run? (I have no clue and just throwing that idea out there).

Another issue is that she's pushing to use my GC instead of hers (cause hey, why use her GC when she can inconveniance someone else!). I keep pushing to not do this because I'm in the middle of weekly analysis myself, and don't really want to be messing around with my GC and potentially messing it up (I'm pretty sure Murphy's ghost hangs around here, if it can go wrong it does). Now her idea is to do it on the weekends when I'm not here (which depending what we end up with column wise, is unrealistic due to the # of samples she has). So this would mean shutting my GC down, letting it cool, switch columns, get it back up, let it re-stabilize, do her analysis, then shut down, switch it all back so I could do my analysis the next week. And she's wanting to do this on a weekly basis for like the next month and a half. So, any input on this - bad idea? or just me being territorial, lol

Thanks!
Jennifer

[muGC]

1) A molecular sieve column will separate O2 and N2, and be a fast analysis.

2) Depends on how important your analysis is to you and your customers. Can it handle being down because of trouble switching methods? I prefer keeping GCs to one application over long periods of time.

[jearles]

Thanks. I also prefer to keep the GC setup the same as much as possible to avoid trouble. I went ahead told her I didn't want to try this, and if it opens up she's welcome to use it then. I can't really take the risk of an extensive shut down at this point.

I will go ahead and suggest ordering a MS in case the opportunity arises to run her samples, since they're not particularly expensive and shouldn't go to waste (I know of one other person who's had interest before in O2 analysis, so it may come in handy later on). Restek suggested a MS, though we may have to do a bakeout periodically depending on how much moisture and CO2 is in her samples. Even with that I think its a much more practical solution.

[AICMM]

Cutting down another Hayesep DB is not an option, physically it would be very difficult and the separation would suffer as well. On a DB the nitrogen should elute before the O2 so you might have an issue on your hands there as well. Sieve is the answer, not that expensive, very robust, not too difficult to install and certainly bakeable. With a TCD and a few hours of work you should be ready to go.

Best regards,

AICMM

[aidnai]

I'll add my voice to the consensus. Molsieve/TCD, separate GC.

Note that air contains around 1% argon. You will have to decide if this is relevant in your situation. A typical molsiv 5a column will not separate argon and oxygen. Low temperature, 20-30m capillary column can get you there, as long as the peaks aren't so big as to mask each other.

[muGC]

By low, aidnai means the use of cryo to get below zero C in the GC.

[AICMM]

By low you may also mean concentration. I've seen PPM O2 and argon separated on mol-sieve, not cryo.

Best regards,

AICMM

[muGC]

By low you may also mean concentration. I've seen PPM O2 and argon separated on mol-sieve, not cryo.

You're right, on a PLOT mol-sieve column, yes. Not on a packed mol-sieve, you need cryo unless that column is very long!