Chromatography Forum

I have no experience with GC and need advice for testing a sample of fragrance oil used for candles. The fragrance is vanilla, but one sample smells "off" so I need to determine why this fragrance is bad compared to a standard sample of the fragrance. What should I be looking for? How significant is the area under the peaks.

I have one word of advice: "outsource".

I'm sorry if that sounds harsh, but the fact is that flavors/fragrances chromatograms are extremely complex. What you are proposing to do is the equivalent of asking a 16-year-old who has just finished driver's ed to go out and win a NASCAR race.

HS-SPME (Headspace Solid Phase MicroExtraction) - GC could be a method for determining flavors and fragrances from solid samples. Engewald et al. (University of Leipzig) did it some years ago.

Hi Becca

I have to agree with Tom. With just a little bit of knowledge about how to run the instrument you will be able to generate chromatograms of your samples, but what do you do then ?

Suppose the chromatograms look different - is that because the samples are different, or due to analytical variability ?, and if it is because the samples are different do the differences have anything to do with the difference in quality ?

Suppose the chromatograms look the same - this could just mean that there is a contaminant that is below your detection limit, or buried under the peak of a component that is supposed to be there (you will get a huge peak for vanillin that will obscure all sorts of other "interesting" things).

GC is a remarkably robust and forgiving technique in that you nearly always get a chromatogram when you inject a sample - it is the optimisation and interpretation that presents the challenge. Aroma and flavour troubleshooting needs a range of sample prep techniques, optimised chromatography, mass spectrometry and maybe 2-dimensional GC and GC-sniffing, not to mention a good nose ! It's fun, but difficult !

Regards Peter

Becca - I work for a large consumer products company, and, yes, we even do have candles. The way I read your question is that you feel you have an off-odor fragrance, and would like to compare to the retain or standard. OK, if that's what you want to do, I'll first tell you what my company would do: simply reject it, as it's off-odor, and that's the supplier's problem, not yours, and it's also not your problem to identify the cause.

That said, we oftentimes do investigate fragrances to find reason for off-odor; we dilute the fragrance oil 1:10 in methanol, use GCMS withsplit injection, starting at 50C, ramping slowly to about 240C. We then overlay/stack and compare suspect fragrance oil to standard, and there can be 75 or more components; our in-house GCMS spectral library enables us to identify components positively. GC-FID can be used similarly, but that doesn't identify the components readily. Yes, we've seen suppliers make errors, such as wrong components or left-over fragrance contamination (this is why GCMS is so valuable, can positively show the non-desired peaks were fragrance components, just had a case like that). If the candles are already made, you're most likely SOL. To compare candles for fragrance, weigh candle into a vial, add solvent such as methanol or DMD, cap securely, heat to melt the candle, shake well, etc., to mix, allow sample to cool so the paraffins kick out of solution, leaving the fragrance soluble in the solvent, then filter and inject that.