News

Authentication of Extra Virgin Olive Oil (EVOO) sometimes requires a panel of different tests and – with more sophisticated adulteration – a weight of evidence interpretation.  For more crude adulterations a single test is often enough.

One of the available tests is for fatty acids ethyl esters (FAEE).  These are more concentrated in lower quality oils (e.g.improperly stored or overripe), formed from ethanol which is a result of fermentation. EU legislation specifies a maximum 35 mg per kg FAEE concentration in EVOO.

FAEE concentration is officially measured using gas chromatography (GC) after recovery by silica gel column chromatography. While highly accurate, this method is complex, time-consuming, and relatively expensive.

This paper (purchase required) reports an alternative approach to FAEE measurement by using infra-red spectroscopy (FT-IR) with machine learning. A dataset of 170 olive oil samples with FAEE concentrations ranging from 1.81 mg/kg to 109.00 mg/kg were analysed using FTIR. Spectral data were preprocessed and used to train various regression models.

The authors report that the best performance was obtained with an XGBoost model. Explainable AI techniques (SHAP) enabled interpretation of the model and identification of spectral regions mostly associated with FAEE content.

They conclude that combining FT-IR spectroscopy with advanced ML models—particularly XGBoost—can effectively predict the concentration of FAEE.

Photo by Massimo Adami on Unsplash

Food fraud prevention and detection priorities can be different in different countries.  In Iran, as in many countries, pork meat is an unlikely adulterant in beef or chicken sausages as there is virtually no pork production; it is legally and culturally proscribed.  Donkey and horse, however, is not food grade but is cheap and readily available as an adulterant.  Laboratories with PCR are scarce and the need is for rapid, portable verification tests.

The researchers in this study (open access) sought to address this need by developing a classification model using non-destructive FTIR.  They deliberately omitted any extraction or defatting step so that the test could be applied directly to a 3mm slice of the intact sample.  They trained the model using sausages prepared in-house that mimicked – as far as possible – the typical recipes used in Iran (40 – 60% meat content, along with soy flour, egg, herbs and spices).  They prepared 20-each of beef, chicken, horse and donkey sausages using meat sourced directly from veterinary schools.  For the training set, triplicate sub-samples were measured from each sausage and then the triplicates averaged.  Some pre-processing was applied to the data before dimension reductions using supervised machine learning.  30% of the samples were reserved as a validation set and kept independent of the training set.  Additionally, within the training set, a 5-fold cross-validation procedure was used to iterate an internal check against over-fitting.

The researchers were able to separate the four species into distinct clusters using Principle Component Analysis.  They also postulate a chemical rationale as to why these identified signals should differ between species.  They conclude that their approach could form the basis of a rapid non-destructive test with practical application.

Image – from paper 

This paper (open access) reports the construction of a classification model to detect the adulteration of white pepper with mung bean flour utilizing Fourier Transform Infrared (FTIR) spectroscopy combined with chemometric techniques.

The authors prepared their own reference samples in-house by grinding locally sourced white pepper (Malaysian origin) with bean flour ranging from 3 – 50%.

They report that adulterants can be detected even at the lowest concentration prepared using the Partial Least Squares (PLS) method and chemometrics.. The second derivative FTIR spectrum in the range of 3712-650 cm⁻¹ was identified as the optimal calibration model.  The PLS Discriminant Analysis (PLS-DA) method also successfully classified pure white pepper samples from those adulterated with various concentrations of mung bean flour.

Cocoa is high on many companies’ current risk radar for authenticity threats, due to recent supply pressures and price increases. Carob has legitimate uses as a cocoa replacement, and carob flour has been cited as a potential cocoa adulterant.

 A number of chemometric classification methods to differentiate cocoa from carob recently have been proposed, including one featured in our blogs in January based on DART-MS.  In a more recent publication (purchase required) the authors use the alternate method of near and mid-infrared spectroscopy before applying various chemometric approaches.

Spectral data were collected using four different infrared spectrometers: a benchtop FT-NIR system, two portable NIR instruments, and a benchtop FT-MIR-ATR. Reference samples included pure cocoa, pure carob, and their mixtures with carob concentrations ranging from 0 % to 60 %. Both classification and regression models were developed to detect and quantify the presence of carobs in cocoa powder. Classification models, including Random Forest (RF), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), k-Nearest Neighbors (kNN), Linear Discriminant Analysis (LDA), and Soft Voting Classifiers, demonstrated superior performance for discriminating between cocoa powder, carob powder, and cocoa-carob mixtures, particularly using the benchtop FT-NIR. Similarly, regression models - RF, SVM, MLP, kNN, Partial Least Squares Regression, and Voting Regressor- exhibited robust predictive capabilities, particularly, FT-MIR and portable NIR.

Overall, these findings highlight and prove the potential of NIR and MIR spectroscopy as rapid, robust, and non-destructive tools for screening and quality control in food authentication.

Photo by lindsay Cotter on Unsplash