Can the opinion of sensory panels be reduced to chemical analyses?

Skip to Navigation


  • Published: Nov 1, 2016
  • Author: Ryan De Vooght-Johnson
  • Channels: Laboratory Informatics / Chemometrics & Informatics
thumbnail image: Can the opinion of sensory panels be reduced to chemical analyses?

Acquired taste

Researchers in Germany have developed a using GC-MS and olfactometry combined with linear discriminant analysis method to classify samples of virgin rapeseed oil based on sensory qualities.

Swilling water across their palates, priming their acute gustatory cells for the work ahead, three members of the German Society for Fat Science have convened.

“Musty…rancid even,” you may hear one cry disapprovingly, whilst jotting a mark next to ‘improper storage’.

Across the room, a fellow adjudicator disagrees, drawing the democratic decision to a tie with a swift stroke of a pencil. The third sits on the fence; neither here nor there.

Next up: a virgin rapeseed oil from central Italy. The panel is unanimously impressed; awarding it threes and ones out of five for the desirable attributes of ‘seed-like’ and ‘nuttiness’ (perfect scores for any aspiring green rapeseed oil). The night draws on and tiredness enters, even for trained experts, but the arduous task of scoring must continue—there are still twelve more oils to be swished and wafted.

‘Sensory analysis by a trained sensory panel is very time, labor, and personnel intensive and the results sometimes entail uncertainty,’ claims Anja Bonte, a researcher from the Federal Research Institute of Nutrition and Food in Germany.

Molecular sensing

Writing in the European Journal of Lipid Science and Technology, Bonte wanted to get to the heart of the matter: what are the chemicals that dictate whether a virgin rapeseed oil is good or bad? ‘It is known that the characteristic sensory perception can be explained partially by the occurrence of volatile, aroma active minor compounds that are formed during cultivation of the fruits or processing and storage of the oil,’ she explained.

In order to answer this question, Bonte purchased 79 virgin rapeseed oils directly from their source and enlisted five expert tasters to grade these out of five as per the strength of defective attributes: whether fusty, musty, roasted, burnt or rancid, for example. Next, Bonte analysed these oils for their volatile components by dynamic headspace GC-MS.

In parallel, she also assayed distinctively good and distinctively bad oil samples by so-called ‘GC-MS-olfactometry’ in which five experts judged the odours of each major elution peak. The compounds underlying these peaks were then identified from a mass spectral library.

Blind classification

Using principal component analysis, the German researchers teased out patterns in 13 major volatile components contained within 43 sensory-good or sensory-bad oils. Of these, 3-methylbutanal and 2-methylpropanal—two entities that emerge from the remnants of degraded microbes—were most prominent, suggesting that oil spoils mainly because of biological chemistry.

Furthermore, what was particularly interesting was that good oils were not sensory pleasant because of the presence of definitive volatile compounds, but rather that they were pleasant because of the absence of notable volatile entities. ‘The impairment of the sensory quality of virgin rapeseed oil was not the result of the disappearance of volatile compounds,’ the paper clarifies, ‘but due to the appearance of degradation products typical for sensory bad oils.’

Wrapping up their comprehensive study, Bonte and co. let their classification strategy loose on an independent set of 13 oils, which were ‘satisfactorily’ categorised in accordance with experts’ opinion.

Though the authors acknowledge that their strategy should not replace trained experts altogether, they suggest that it could ‘help to reduce the number of samples which have to be tested by the group’ and could therefore ‘save a lot of man power’.

Related Links

Eur. J. Lipid Sci. Tech., 2016, Early View paper. Bonte et al. A chemometric approach for the differentiation of sensory good and bad (musty/fusty) virgin rapeseed oils on basis of selected volatile compounds analyzed by dynamic headspace GC-MS.

German Society for Fat Science

Article by Ryan De Vooght-Johnson

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.

Follow us on Twitter!

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Most Viewed

Copyright Information

Interested in separation science? Visit our sister site

Copyright © 2018 John Wiley & Sons, Inc. All Rights Reserved