Chromatography Forum

Dear All,
I am looking for you guys help. I am developing method for cholesterol analysis in food. I tried to run Cholesterol standards to get calibration curve but I am not getting any cholesterol peak except solvent peak.
I am using following system.
Varian GC 3900 ith FID detector
J& W DB1 Column 30 meters length, 0.32 mm I.D, 0.25 umFILM.

I am using following analytical conditions.
Injector Tempertaure 250 C.
FID Temp 300 C
Column 190 C , hold 2 minutes, Increase 20C/min to 230 C and hold 3 minutes, incresae 40 C/min to 255 C and hold 25 minutes.
H2 as carrier gas 2 ml/minutes,
Nitrogen make up gas with air.

I am injecting 1 microliter of the working standard, but only get solvent peak around 2 minutes and after that nothing. Should I condition the column, while column manufacture has alrtaedy stated taht it is pre conditioned column.
Should I change the liner?
Please guide me . Thanks for your time and help.

Regards
Atique

cholesterol is a free alcohol and is prone to decomposition if the injector port and column are extremely deactivated. Also if injector temperatures are too high they will not remain stable.

Since you are only going to 255C I doubt the cholesterol peak will elute at that low a temperature.

I would lower the injector to 220 or 200 and increase the oven final temperature to 330C.

I would also suggest a shorter column. Try a 5 or 10 meter column.

Good luck with free alcohol. TMS derivative is recommended.

Rod

We've used the TMS derivative here.

Cholesterol needs high temperatures or very long times to elute from non-polar columns. I get it off a Restek wax column at 240C after a longish temperature programme (there are lots of other things that I am also interested in).

Conditioning the column is unlikely to make any difference at all.

One thing that you should check (and then check again) is that the concentration of the solution that you are injecting puts 50 - 100 ng of cholesterol onto the column. This is plenty to give you a nice big peak as long as everything is working right.

Peter