Chromatography Forum

I folks Once again, I will preface this by saying that I am very much an amateur at GC. Now, here is what I am doing:
I am pulling gas samples directly from a tank that is a blend of N2/CO2 at ratio of 60/40 (analyzed by supplier). I monitor the GC inlet pressure and keep it fixed via a regulator. I also monitor the GC outlet flowrate and keep it fixed with a rotometer.

I analyzed the blend several times to get consistent retention times: t_N2 = 0.737 and t_CO2 = 1.911.

Now I disconnect the line at the tank of N2/CO2 and connect it to a different blend of N2/CO2 at 86/14. I fix the inlet pressure and outlet flow to the same exact values as the previous blend. However, this time I consistently get t_N2 = 0.725 and t_CO2 = 1.955.

These differences are significant enough that if I calibrate the method using one blend, it cannot recognize the other blend as being N2 and CO2. Furthermore, if I connect pure CO2 I get yet another retention time.

I know that there is so much info that goes into a method, that you probably need a lot more info to pinpoint the problem, but maybe you guys can give me some places to start since I really don't know much about this. I am under the impression though that I should be able to calibrate the method without having to know ahead of time what the mixture composition is. That would just be silly.

Thanks for reading

EDIT: How the heck did I end up in Data SYstems forum? Sorry Maybe a staff member can move this to GC?

I would be willing to bet your peaks are not perfectly symmetrical.

I would also bet that since you are measuring such high percentages, your column is overloaded or close to it.

I would also suspect that you are doing gas solid chromatography, not gas liquid chromatography.

Given the above your peaks change shape with their concentration, ie, the little points at the top of the peaks are not always in the same position relative to the base of the peak.

Thus your retention times change slightly.

Your peak recognition parameters are set too tightly.

Solution:

widen the parameters.

Problem solved.

best wishes,

Rod

I would be willing to bet your peaks are not perfectly symmetrical.

I would also bet that since you are measuring such high percentages, your column is overloaded or close to it.

I would also suspect that you are doing gas solid chromatography, not gas liquid chromatography.

Given the above your peaks change shape with their concentration, ie, the little points at the top of the peaks are not always in the same position relative to the base of the peak.

Thus your retention times change slightly.

Your peak recognition parameters are set too tightly.

Solution:

widen the parameters.

Problem solved.

best wishes,

Rod

Hello Rod

You are correct with all of your assumptions. And I relive that your solution will work. Unfortunately I am too stupid to figure out how to 'widen the parameters.' I thought that I could figure it out, but the parameters that I changed do not seem to affect the results. I know is a nuisance to ask someone to hold my hand through this, but I would be grateful. I believe I am almost there, I just need a little more guidance.

Here is some more info that should help to troubleshoot:

When I calibrate the method with the N2/CO2 60/40 blend, the retention time that I have put in the peak table for each are shown in the figure. You can also see that I adjusted the retention time parameters 'Define Peak Windows' to 10% thinking that would solve the problem (it did not):




When I run an analysis on the other blend N2/CO2 86/14 these are the results that I get:



As you can see, the N2 result has the warning 'C = Out of Calibration range'

What is strange is that the deviation of the N2 time and CO2 time from the calibrated time is ~3% for both cases. So I do not know why it is kicking out a warning on N2 but not on CO2. This lead me to believe that I need to look in the method to see what is different about the parameters on CO2 from those of the N2, but I am still coming up short.

I also set the following tolerances pretty wide under the 'Calibration Setup' tab:



and this was already set under 'Calibration Verification' tab



I would be grateful for any more tips as to how to remedy this. If it helps, I am using Varian 3800/Star WS 6.2.

Thank you for your time and patience

Since I have not had experience with your software for many many years, perhaps another poster can come to assist you. Having experience with too many software packages I now get them confused. I would not wish to misspeak.

Otherwise, I can only suggest you call your vendor or someone in your local Chromatography Group of the ACS.

best wishes,

Rod

Since I have not had experience with your software for many many years, perhaps another poster can come to assist you. Having experience with too many software packages I now get them confused. I would not wish to misspeak.

Otherwise, I can only suggest you call your vendor or someone in your local Chromatography Group of the ACS.

best wishes,

Rod

Thanks for your help Rod! I'll call Agilent on Monday and see what I can dig up.

Damn Rod, that was some premium troubleshooting right there. Right to the point .

Thanks.

Even us old timers (like a broken clock) are correct twice a day.

Rod

Out of Calibration range - this is probably another problem (note that your peak is correctly identified after increasing the peak windows). I will suspect this to be related to the quantification - probably results which are higher that the highest standard (i.e extrapolated) are marked this way?

Probably your software manual has a troubleshooting section or list of error messages?