Quantitative gas chromatography–olfactometry. Analytical characteristics of a panel of judges using a simple quantitative scale as gas chromatography detector

Some questions related to the ability of gas chromatography–olfactometry (GC–O) to provide quantitative measurements of the concentration of a given odorant in an extract are explored and discussed. A panel of eight individuals has been used to evaluate the intensity of 15 odorants present at different concentrations. The use of very simple scales, such as a 0–3 scale, makes it possible to build calibration graphs based on the different stimulus–response models (Fechner, Stevens, Hill) and, with a proper calibration, up to nine different concentration levels can be discriminated by an eight-judge panel. The signal shows a good long-term stability, and its precision varies between 3.7 and 8% of the whole scale, with 5.7% as average. Sensitivity is extremely dependent on the compound: in the best case a concentration change of 20% can be detected, while in the worst, concentrations must differ more than one order of magnitude. In average, concentrations must differ between 2 and 4.7 times (including calibration error) or between 1.2 and 2.3 (excluding calibration error). The performance of the different judges, the effect of the close elution of two odors and the benefits derived from the use of more complex scales (7-points) are briefly discussed from the perspective of the analytical performance of GC–O methods.