Odor - The Electronic Nose Artificial Olfaction Technology - page 97

Table 4.5 Relationship
between odor intensity and
odor index

Odor intensity Range of odor index

2.5 10-15

3.0 12-18

3.5 14-21

Odor classification is studied using psychophysical methods would suggest that
there are several primary odor groups. These include camphor-like, musky, floral,
pepperminty, ether-like, pungent, and putrid. Unique mixtures of aromatic com-
pounds including phenolics, ketones, and terpenes [ 5 ] determine the characteristic
odors of fruits and vegetables. Smell sensitivity is genetically determined and
varies from person-to-person, hence different people perceive or may react to
different odors, or react differently to the taste of foods.

4.7.1 Odor Index

The odor index is defined as the dimensionless ratio of the vapor pressure and the
100 % odor recognition threshold (the concentration at which 100 % of the odor
panel detect/recognized the odor as being representative of the odorant being
studied). The odor index provides information on the potential of a particular
compound to cause odor problems under evaporative conditions [ 19 ].
The ''Odor Index'' is used in Japan to quantify the intensity of odors [ 19 ]. The
odor index is equal to ten times the log of the odor concentration (i.e., Odor
Index = 10 9 log(Odor Concentration). Table 4.5 shows the relationship between
odor intensity and odor index. The odor concentration is measured using the
Triangular Odor Bag Method. The odor indices of the common odorous com-
pounds are given in Table 4.6 . Also the variance of odor measurement illustrates
in Fig. 4.8 .

4.8 Odor Analysis Techniques

Conflicting criteria must be considered when choosing an odor assessment.
For example:
• Timing of Data Collection—Must data be collected on-line? If not, how soon a
sample is collected, should the analytical results be available?
• Sensitivity, Selectivity, and Stability—What range of analyte concentration or
property value can be expected? What are the possible interferences or matrix
effects? How important is day-to-day reproducibility? How important is

instrument-to-instrument reproducibility?