The Coi file for the commercial classification of olive oils reports very few parameters: defects, fruity, bitter, hot. Several recent researches have classified the compounds which are responsible for the olive oil aroma.
For instance, it has been shown that the mold defect is strictly correlated with 1-octene, isopentanol, alpha-copaen. The wine like defect is correlated with ethyl acetate. The sludge defect is correlated with propanol-1, ethyl pentanoate, methyl butyrate, and ethyl propionate. The rancid like defect is correlated with butyl butyrate, 1-pentanol and butanal. The RISCALDO defect is associated to butyl acetate, 2-butanol and ethyl butyrate.
As regards the fruity aroma, molecules which generate it include 1-penten-3-one, 2-4-hexadienal, 1-penten-3-oi.
The bitter is due to phenolic substances, such as the dialdehydic form of decarboxymethyl oleuropein aglycone, the aldehydic and hydroxylic forms of oleuropein aglycone, oleuropein, tyrosyl acetate.
The hot flavor is correlated with the dialdehydic form of decarboxymethyl ligstroside.
The scientific papers which proposed these identifications can be further refined and better markers can be identified, with higher correlation coefficients with the different oil aromas. Still, while technical and scientific improvements are being developed, these indices have already been validated by a parallel comparison with an expert panel.
What is the missing step before the PNALE is replaced by a HPLC analytic exam to carry out commercial classification?
The repeatability and the reliability of the chemical instrumentation are not the issue, of course. Two are the main problems.
The quantitative threshold for single molecules to identify a defect (or a quality) is defined by the perception threshold of prepared and qualified tasters.
In a complex matrix such as an extra virgin oil, a single marker can not be enough to guarantee a good sensitivity. The interaction and simultaneous presence of several molecules which concur to generate the same sensation may lead to an organoleptic positive test although this does not necessarily mean that a marker has necessarily passed the threshold.
The other issue is that the commercial classification is based on the organoleptic discovery of defects. Still, it is known that certain features of oils, such as a very intense fruity flavor, can mask a slight defect from the taster. Is this ok to have an oil flunk a test because a mathematical model says that a defect should have been detected but was not because the tasters’ senses were confused by other sensations? In other words, the defect is a defect when it can be detected or just when it is present?
While the first issue, relevant to the applicability of the analytical method, can be probably solved by further research, the second one is more of a cultural consideration which requires a deeper discussion.
Tasters, though, do not have to be afraid of these systems, since they can be relieved of routine tasks like the commercial classification, while they could dedicate more effort to a more complete and emotional organoleptic description, something that cannot be done by any machine or analysis.