Essential Oils sample prep

[m.finnBMBM]

What is a good sample prep for essential oils for GC/MS?

[Don_Hilton]

It depends a bit on what you are doing. I've injected them neat (1 or 0.2 microliter injections) and have obtained good results in looking for various trace level compunds and for profiling the oils.

Note added in subsequent edit: It also depends on the oil in question. Some oils are not suitable for injection into the GC inlet.

[HiggsBoson]

I have allready do by direct injection, by just with dilution 1/10 or 1/100 it depends on what you are chearching for.

You can do a phase separation too, (liquide/liquide separation) after you can do SPE purification.
It depends of your need.

[m.finnBMBM]

It is a coffee oil extract. I have tried a 1:25 dilution in hexane, splitless, 1 uL and there seems to be quite a few trace components that aren't showing through enough. But there are a few larger peaks much later on. I tried a 1:5 with a 1:20 split, same thing. The large peaks are overloaded. I may try a 0.2 uL injection, 1:10 (I don't want to change the liner), with the 1:5 dilution. I have a cyclo double gooseneck liner; is it a problem to run splitless with this liner? Do you have any suggestions on the thought process to choosing parameters?

[Don_Hilton]

One problem with essential oils is the wide conentration range in the samples. If you need a lot of material on the column to see the smaller peaks and that gives rise to overloading of the column for the larger peaks, the injection is not the problem - it is what the column can take. When you make the big peaks smaller, you make the small peaks smaller as well. Can you go to a column with more stationary phase to handle the larger components and still see the smaller components? It may be that a longer column may help to obtain better resolution. (But too much column and the small peaks broaden and get lost.)

So the question is what column are you using? And, do you know what the larger components are to know if they are compatible with the column?

And, on a side note, I like GCxGC for this kind of work because mixtures like coffee oil will have many many many peaks - and in a one dimensional separation it is very difficult to pull them all apart.

[m.finnBMBM]

> Don_Hilton

I am using a Rtx-1 60 m, 0.25mm, 0.25um. What is the best balance for seeing trace components of let's say clove oil while not overloading the column? What sort of damage is done to the column by overloading, and what can I do to render these problems?

[Don_Hilton]

The balance depends on what you are looking for. I have run things like clove oil and have let the eugenol overload the column - because the eugenol was not what I was interested in. Overloading a column does not hurt a column. We do that with solvents we use to dilute with samples all the time. The overload condition gives the same issues as a large sovent peak. It distorts peakshapes of coeluting compunds and it overloads detectors. With a mass spectrometer, this becomes a problem when there are too many gas molecules in the ion source -- you burn up filaments and age detectors prematurely.

Without more specifics of what you are trying to do, I can only suggest: Make a few injections under varying conditons and see what you can see.

Also, consider: Can you run on a shorter column? You have many theoretical plates on a 60 M column. But it takes a longer time to get the compunds out, so the peaks become flatter - and harder to detect. If you can get the job done wiht a 20 M column, it has advantates in peak sharpness and cycle time on your analyses.

[m.finnBMBM]

I am simply screening samples to see what we have. Thank you for all of your help.