peak area variability

[John Keck]

I am performing an analysis for PPM levels of 1-propanol (retention time 12 min) in 40% ethanol solutions. I am using capillary GC split (60:1) injection. FID detector. I use 3-pentanol (retention time 14 min) as internal standard. My problem has to do with peak areas. The areas vary a lot from run to run. The area might be 10 units for one run and 1 unit for a duplicate injection of the same sample.

[chromatographer1]

John,

You stated 40% alcohol, does that mean 60% water?

if that is the case you have a difficult situation as water can give serious problems to reproducible injections as you have discovered.

There are issues of flooding the column surface and the injection liner surfaces with water. One would need extremely consistent splitting without discrimination so liner type and column tip position is very critical.

Minimize your injection size and optimize these other factors and you should get suitable numbers but I don't think it will be an easy task.

It is much easier to use direct injection than split injection in these cases.

Multidimensional chromatography (not GC X GC) using heartcut technology makes this a breeze.

If you need more assistance you are welcome to call Supelco technical service, the help is free.

best wishes,

Rod

[Peter Apps]

Hi John

You have chosen a tricky sample !!

If you have an equilibrium headspace sampler I would try that rather than split injections.

If you are stuck with a split splitless inlet then direct injection (which needs only a specific inlet liner and probably a retention gap) is first choice. At ppm levels you could go to splitless injections without overloading the column with analyte, you will need a retention gap with a polar deactivation. Indeed, ppm split 60:1 must be giving tiny peaks and a good deal of the variation in area might just be because the peaks are small relative to baseline noise.

You will need to play with inlet temperature, column programme start temperature, and injection parameters such as injection speed and pre and post injection dwell time. If you can post some details of your hardware we might be able to make some specific suggestions.

Good luck !

Peter