ISSUE 03 | JULY 2016
Food Quality Magazine
olive oil harvest in Italy, during which
222,000 metric tons of valuable
Italian olive oil was produced. In
the same period, the Italians themselves consumed 521,000 MT, and
still managed to export 208,000 MT
(Figure 6). It is very obvious that the
figures do not tally. One may say that
the difference is explained through
imports. However, in that period the
Italians imported less than 100,000
MT of olive oil. So there is a significant discrepancy, which may be the
result of fraud in the form of mislabeling or the addition of lower quality
oils. Price surges during periods of
low supply give further incentive to
fraudsters. As it can be clearly seen
in Figure 7, during the poor harvest
season 2014/2015, the prices for
olive oil increased significantly.
Detection Methods
There are still a number of fraudulent
food ingredients which can be detected by conventional technologies
and even using household tools.
Sawdust added to spices, red clay
or brick dust added to paprika, and
yellow clay added to turmeric can
be identified with a common light
microscope. The addition of starch
to spices or milk can be identified by
adding iodine; starch is present if the
resulting solution turns blue. Even
the addition of dried and ground cow
dung to some ground spices can be
detected using a glass of lukewarm
water – you will literally smell the
difference.
However, with food fraud being a
very lucrative enterprise, fraudsters
often go to the extremes to fake
their products with more and more
sophisticated methods. How can
these sophisticated frauds be uncovered? With even more sophisticated analytical methods. For fluid
milk, the deployment of targeted
and non-targeted mid-Infrared (IR)
spectrometry presented a challenge
for fraudsters as anything from melamine to urea added at economically
motivating levels could be detected20.
Figure 6. Production, Italian consumption and export of Italian olive oil
2014/2015
Figure 7. Production vs. price for extra virgin olive oil over time
More recent developments have
coupled these “non-targeted” approaches with high resolution mass
spectrometry. Non-targeted analytical approaches involve the combination of lab-based methods with
statistical software to determine if
a sample falls within an expected
set of parameters. This technology
is very sensitive and enables detection of a wide range of adulterants as well as the identification of
even small quantities of adulterants.
It typically works using a learning
approach: the system is trained using
authentic reference samples as well
as adulterated samples. Measuring
a multitude of parameters, a profile
Heggum, C., & Holroyd, S. (2011). Integrated Supply Chain Management (Publication No. 449). Retrieved May 31,
2016, from http://www.fil-idf.org/Public/PublicationsPage.php?ID=27121#list
20
10