GC/MS issue

[BGDY]

Hi, Guys

We have some issue with our GC/MC and I need help. We are using this equipment to determinate the % of menthol in our samples, but when I am making two or more injections from exactly the same vial the % difference between them is huge. Do you have some idea how I can fix this issue. We are using the correct area to calculate our %. Thank you in advance.

[chemtexpps]

Better to use FID than MSD. But i do not know How you are calculating RPD using R.T. use concentration to calculate RPD.

[BGDY]

Yes, we are using the Correct Area to calculate our % and at the end we have these differences. I was thinking that probably the volume of the injections is not the same all the time. How I can check this ? Actually, we have these issues only with the menthol and winsense 500 (sometimes). I am looking for a possibility reasons.

[chemtexpps]

Yes, we are using the Correct Area to calculate our % and at the end we have these differences. I was thinking that probably the volume of the injections is not the same all the time. How I can check this ? Actually, we have these issues only with the menthol and winsense 500 (sometimes). I am looking for a possibility reasons.

Yes you have to inject exact amount each time to compare .

[chemtexpps]

Yes, we are using the Correct Area to calculate our % and at the end we have these differences. I was thinking that probably the volume of the injections is not the same all the time. How I can check this ? Actually, we have these issues only with the menthol and winsense 500 (sometimes). I am looking for a possibility reasons.

Yes you have to inject exact amount each time to compare .

R U using auto sampler or manual injection?.

[BGDY]

Auto Sampler and now I am running some L- Menthol and I can see that I have some strange results and this is only the beginning.
I mean this :
656ppm- 102671554 (the correct area)
328ppm- 37705809 (the correct area), this one should be on the half of the first one like correct area ?!

[Consumer Products Guy]

I don't know if this will help, but we always use trimethylsilyl derivatization when we do menthol, and use a DB-5 or DB-1 type capillary. We typicaly derivatize most analytes containing hyroxy groups (unless the molecules are real small and polar like small alcohols), the TMS derivatives for some of these seem to be more stable in the GC inlets, so precision is better.

We use GC-FID more often than GC-MS for this.

[BGDY]

Thank you for your advice, but we have only GC/MS. I am confuse because sometimes the GC is working fine with this method, but the most recent runs ,some of them, they were really very ridiculous like results.
How often we have to have a service on the equipment to be sure that everything is fine with it?
Would you share some information for your GC/MS method if this is possible I will appreciate this?!

[AICMM]

BGDY,

I am a bit confused. You are doing percent menthol but you are using a ppm standard? This means there is a big dilution somewhere in your prep? Could this be the source of the problem?

Also, what does the spectrum of the 656ppm look like (exactly the same as 328 ppm?) and how much are you shooting and are your running split or splitless? 650 ppm splitless could be saturating the EM and if you are running split that could be an issue for you.

Best regards,

AICMM

[bhuvfe]

Maybe you can try an alkane (tetradecane or similar) and see if that one is reproducible.
If yes, you will exclude at least your autosampler as source of the problem.

[Peter Apps]

Variations in injection volume can be corrected by using an internal standard. Use a compound with a boiling point close to menthol's, and preferably as close in structure as you can get it. It must separate from other components in the sample (unless you can use selective ions from the mass spectrum). Calibrate % menthol against the menthol:standard peak area ratio.

Peter

[Don_Hilton]

If it helps, a lab I worked in used n-octanol as the internal standard for menthol measurement. The solven used for extraction and method of extraction can be important if you have to extract the sample away from a matrix. We did not derivatize the menthol. And we were looking at probably hundreds of parts per million in the soloution injected into the GC. (We used FID for detection)

[BGDY]

We are using methanol for the extraction. This is the perfect one for my matrix. I'm doing a standard curve usually with 6 points , often I use only 4 points to be in 0.999 for the Linear regression (R2). I don't think that the concentration is the issue, I am making the sample concentration based on my curve ( usually I want to be in the middle, because the mistake will be less).
Probably,we have this issue because we don't make derivatization on the menthol?!

[Don_Hilton]

Derivatization in methanol will be a problem - as it is an alcohol, as is menthol. And you can do chromatography of menthol without derivitization. The next question is how do the peaks look and how does the integration look. For the same sample shot twice, do the integration markers (start and end of peak) fall in the same place along the chromatogram?

[BGDY]

Yes, you are right for the methanol. Probably I have to dissolve the sample in water or some other solvent and after that to make extraction with hexane.
I need really help with this issue I need reproducibility.
About the peaks ,when the volume per injection is 2ul instead 1ul , I have 3 peaks and when I checked them they were all menthol. The curve was fine like R2(0.9991) and I used 5 points but the results were terrible.
I don't think that the autosempler is issue , because I am running fragrances and they are fine like reproducibility. I forgot to mention that I am talking about for L- Menthol.