Wine Chemistry and Biochemistry

8E Identification of Impact Odorants of Wines 399

AEDA

Charm

AEDA

Charm

3.25 fruits

6.12 fusel

6.42 honey

R.T. Odor

Dilution 0 Dilution 1

R.T. Odor R.T. Odor

Dilution 2

3.26 fruits 3.25 something

6.42 sweet

15 Time (min)

3.25 6.12 6.42

Dilution 0

Dilution 1

Dilution 2

R^2 = 32 = 9

R^0 = 30 = 1

R^1 = 31 = 3

time

1

3

9

dilution

FD

Fig. 8E.1Scheme showing the basics of AEDA and Charm analysis. Different sequential dilutions
of the sample extract are analyzed in the GC-O system. In AEDA the judge simply marks the
retention times and the odor descriptions. The FD value is Rp, where p is the p-th dilution of
the extract at which the odor was last detected and R is the dilution rate (3 in the figure). In
Charm, he/she presses the space bar of the computer during the odor detection, and the outputs are
combined to form the Charm chromatogram

reproducible than Charm analysis (Debonneville et al., 2002) or some posterior

intensity methods (van Ruth, 2004). However, the dynamic range of the response

is narrower in the case of NIF and this means that in the practice it is easier to dis-

criminate concentration differences in posterior intensity methods (van Ruth, 2004).

8E.2.1.3 Techniques Based on the Measurement of Intensity

The first technique of this type was OSME (McDaniel et al. 1990; Miranda-Lopez

et al. 1992) which was introduced at the beginning of the 1990s. In this technique the

judge handles a knob connected to a variable resistance which is actuated following

the intensity of the odor eluting out of the column. That is, the judge acts as a gen-

uine GC detector. The output is something very similar to a chromatogram (aroma-

gram, odorgram or osmegram) in which the peaks are related to the intensity (peak

height) and intensity×time (peak area) of the odors presented in the sample. The

technique is really appealing; however, in its original form the task imposed on the

judges is extremely difficult and, consequently, the results are poorly reproducible

and difficult to handle. It should be considered that the judges require some time to

detect the odor, to estimate their intensity and to assess simultaneously its quality,

and that a continuous monitoring requires a continuous and systematic feedback

(da Silva et al. 1994). As a consequence, the technique has been simplified so that

the judges do not try to follow closely the odor intensity of the effluent, but rather

to detect the peak start and end and to provide an evaluation of the intensity. The

peaks, then, look like simple triangles instead of genuine chromatographic peaks.