Discrimination of edible olive oils by means of synchronous fluorescence spectroscopy with multivariate data analysis

Detection and quantification of extra virgin olive oil adulteration by other grades of olive oil using front-face fluorescence spectroscopy and different multivariate analysis techniques

This study explores the use of front-face fluorescence spectroscopy to detect extra virgin olive oil (EVOO) adulteration with lower-grade olive oils (virgin, ordinary virgin, lampante virgin, refined, and pomace) at 5-50 % adulteration levels. Emission spectra were analyzed using principal component analysis, factorial discriminant analysis, and partial least square discriminant analysis (PLS-DA) at excitation wavelengths of 270, 290, and 430 nm. PLS-DA at 430 nm provided the best results, achieving 100 % classification accuracy and perfectly separating 12 groups of pure and adulterated samples. For purity prediction, regression models (partial least squares, principal component, and support vector machine) applied to emission spectra data yielded high R2 values of 0.995, 0.96, and 0.98 at 430 nm, 290 nm, and 270 nm, respectively, with a low prediction error of 1.09 %. These findings confirm the method's high accuracy for detecting and quantifying EVOO adulteration.

Excitation wavelength analysis of a laser-induced fluorescence technique for quantification of extra virgin olive oil adulteration

The performance of the laser-induced fluorescence (LIF) technique is greatly affected by the excitation wavelength (EW). This study aims to find an appropriate EW that can be used for analyzing extra virgin olive oil (EVOO) adulteration quantification by comparing the effect of different EWs. The EWs of 405 nm, 450 nm, and 532 nm were selected to perform the comparative experiments. By using the three EWs in the experiments, the LIF spectra of EVOO samples adulterated with peanut oil (PO) or soybean oil (SO) in different proportions, as well as the prediction models established through different multivariate analysis algorithms were analyzed. The linear discriminant analysis (LDA) was applied for qualitative analysis, while the partial least squares regression (PLSR), backpropagation neural network, and k-nearest neighbor were employed for quantitative analysis. The results show that the performance of 450 nm EW is always superior to that of 405 and 532 nm EWs in any model, with a smaller root mean square error (RMSE). Using the LDA-PLSR model, the RMSE is 1.35% for SO adulterants and 1.36% for PO adulterants, respectively.

Feasibility Assessment of Synchronous Fluorescence Spectral Fusion by Application to Argan Oil for Adulteration Analysis

Synchronous fluorescence spectroscopy (SFS) is used for quantitative analysis as well as for qualitative analysis, such as with classification methods. With SFS, determination of a useful wavelength interval between the excitation and emission wavelengths (Δλ) is required. There are a multitude of Δλ intervals that can be evaluated and optimization of the best one is complex. Presented here is a fusion approach for combining Δλ intervals, thereby negating the need to perform the selection by a skilled operator. To demonstrate the feasibility of omitting selection of the best Δλ interval, adulterated argan oil samples are studied. Argan oil is made from the argan tree, endemic to southwestern Morocco, and is well-known for its cosmetic, pharmaceutical, and nutritional applications. It is considered a luxury product and exported from Morocco around the world. Consequently, detection of argan oil adulteration followed by quantitative analysis of the adulterant concentration is important. This study uses fusion of SFS spectra obtained at ten Δλ intervals to first detect adulteration of argan oil by corn oil and then determination of the corn oil content. For detection of adulteration, 15 one-class classification methods were used simultaneously over the ten Δλ sets of SFS spectra. For tuning parameter dependent classifiers such as Mahalanobis distance, non-optimized classifiers are used. Raw classification values are used, removing the need to set classifier-dependent threshold values, albeit, ultimately, a fusion decision rule is needed for classification. For quantitative analysis, two calibration approaches are evaluated with fusion of these ten Δλ SFS spectral data sets. One is multivariate calibration by partial least squares (PLS). The second approach is a univariate calibration process where the SFS spectra are summed over respective SFS spectral ranges, also known as the area under the curve (AUC). For adulteration detection and quantitation of the corn oil, prediction errors decrease with fusion compared to individually using the ten Δλ interval SFS specific data sets. For this argan oil data set, the AUC method generally provides equivalent prediction errors to PLS.

Fluorescent fingerprints of edible oils and biodiesel by means total synchronous fluorescence and Tucker3 modeling

The present work proposes the use of total synchronous fluorescence spectroscopy (TSFS) as a discrimination methodology for fluorescent compounds in edible oils, which are preserved after the transesterification processes in the biodiesel production. In the same way, a similar study is presented to identify fluorophores that do not change in expired vegetal oils, to associate physicochemical parameters to fluorescent measures, as contribution to a fingerprint for increasing the chemical knowledge of these products. The fluorescent fingerprints were obtained by Tucker3 decomposition of a three-way array of the total synchronous fluorescence matrices. This chemometric method presents the ability for modeling non-bilinear data, as Total Synchronous Fluorescence Spectra data, and consists in the decomposition of the three way data arrays (samples_×Δλ_×λ excitation), into four new data matrices: A (scores), B (profile in Δλ mode), C (profile in spectra mode) and G (relationships between A, B and C). In this study, 50 samples of oil from soybean, corn and sunflower seeds before and after its expiration time, as well as 50 biodiesel samples obtained by transesterification of the same oils were measured by TSFS. This study represents an immediate application of chemical fingerprint for the discrimination of non-expired and expired edible oils and biodiesel. This method does not require the use of reagents or laborious procedures for the chemical characterization of samples.

Chemometrics Methods for Specificity, Authenticity and Traceability Analysis of Olive Oils: Principles, Classifications and Applications

Background. Olive oils (OOs) show high chemical variability due to several factors of genetic, environmental and anthropic types. Genetic and environmental factors are responsible for natural compositions and polymorphic diversification resulting in different varietal patterns and phenotypes. Anthropic factors, however, are at the origin of different blends’ preparation leading to normative, labelled or adulterated commercial products. Control of complex OO samples requires their (i) characterization by specific markers; (ii) authentication by fingerprint patterns; and (iii) monitoring by traceability analysis. Methods. These quality control and management aims require the use of several multivariate statistical tools: specificity highlighting requires ordination methods; authentication checking calls for classification and pattern recognition methods; traceability analysis implies the use of network-based approaches able to separate or extract mixed information and memorized signals from complex matrices. Results. This chapter presents a review of different chemometrics methods applied for the control of OO variability from metabolic and physical-chemical measured characteristics. The different chemometrics methods are illustrated by different study cases on monovarietal and blended OO originated from different countries. Conclusion. Chemometrics tools offer multiple ways for quantitative evaluations and qualitative control of complex chemical variability of OO in relation to several intrinsic and extrinsic factors.

Neural networks applied to characterize blends containing refined and extra virgin olive oils

The identification and quantification of binary blends of refined olive oil with four different extra virgin olive oil (EVOO) varietals (Picual, Cornicabra, Hojiblanca and Arbequina) was carried out with a simple method based on combining visible spectroscopy and non-linear artificial neural networks (ANNs). The data obtained from the spectroscopic analysis was treated and prepared to be used as independent variables for a multilayer perceptron (MLP) model. The model was able to perfectly classify the EVOO varietal (100% identification rate), whereas the error for the quantification of EVOO in the mixtures containing between 0% and 20% of refined olive oil, in terms of the mean prediction error (MPE), was 2.14%. These results turn visible spectroscopy and MLP models into a trustworthy, user-friendly, low-cost technique which can be implemented on-line to characterize olive oil mixtures containing refined olive oil and EVOOs.

The effect of thermal treatment on the enhancement of detection of adulteration in extra virgin olive oils by synchronous fluorescence spectroscopy and chemometric analysis

In this study the effect of thermal treatment on the enhancement of synchronous fluorescence spectroscopic method for discrimination and quantification of pure extra virgin olive oil (EVOO) samples from EVOO samples adulterated with refined oil was investigated. Two groups of samples were used. One group was analyzed at room temperature (25 °C) and the other group was thermally treated in a thermostatic water bath at 75 °C for 8h, in contact with air and with light exposure, to favor oxidation. All the samples were then measured with synchronous fluorescence spectroscopy. Synchronous fluorescence spectra were acquired by varying the wavelength in the region from 250 to 720 nm at 20 nm wavelength differential interval of excitation and emission. Pure and adulterated olive oils were discriminated by using partial least-squares discriminant analysis (PLS-DA). It was found that the best PLS-DA models were those built with the difference spectra (75 °C-25 °C), which were able to discriminate pure from adulterated oils at a 2% level of adulteration of refined olive oils. Furthermore, PLS regression models were also built to quantify the level of adulteration. Again, the best model was the one built with the difference spectra, with a prediction error of 3.18% of adulteration.

Identifying and Quantifying Adulterants in Extra Virgin Olive Oil of the Picual Varietal by Absorption Spectroscopy and Nonlinear Modeling

In this research, the detection and quantification of adulterants in one of the most common varieties of extra virgin olive oil (EVOO) have been successfully carried out. Visible absorption information was collected from binary mixtures of Picual EVOO with one of four adulterants: refined olive oil, orujo olive oil, sunflower oil, and corn oil. The data gathered from the absorption spectra were used as input to create an artificial neural network (ANN) model. The designed mathematical tool was able to detect the type of adulterant with an identification rate of 96% and to quantify the volume percentage of EVOO in the samples with a low mean prediction error of 1.2%. These significant results make ANNs coupled with visible spectroscopy a reliable, inexpensive, user-friendly, and real-time method for difficult tasks, given that the matrices of the different adulterated oils are practically alike.

Thermal oxidation process accelerates degradation of the olive oil mixed with sunflower oil and enables its discrimination using synchronous fluorescence spectroscopy and chemometric analysis

We have investigated the effect of thermal treatment on the discrimination of pure extra virgin olive oil (EVOO) samples from EVOO samples adulterated with sunflower oil. Two groups of samples were used. One group was analyzed at room temperature (25°C) and the other group was thermally treated in a thermostatic water bath at 75°C for 8h, in contact with air and with light exposure, to favor oxidation. All samples were then measured with synchronous fluorescence spectroscopy. Fluorescence spectra were acquired by varying the excitation wavelength in the region from 250 to 720nm. In order to optimize the differences between excitation and emission wavelengths, four constant differential wavelengths, i.e., 20nm, 40nm, 60nm and 80nm, were tried. Partial least-squares discriminant analysis (PLS-DA) was used to discriminate between pure and adulterated oils. It was found that the 20nm difference was the optimal, at which the discrimination models showed the best results. The best PLS-DA models were those built with the difference spectra (75-25°C), which were able to discriminate pure from adulterated oils at a 2% level of adulteration. Furthermore, PLS regression models were built to quantify the level of adulteration. Again, the best model was the one built with the difference spectra, with a prediction error of 1.75% of adulteration.

Detection of plant oil addition to cheese by synchronous fluorescence spectroscopy

The fraudulent addition of plant oils during the manufacturing of hard cheeses is a real issue for the dairy industry. Considering the importance of monitoring adulterations of genuine cheeses, the potential of fluorescence spectroscopy for the detection of cheese adulteration with plant oils was investigated. Synchronous fluorescence spectra were collected within the range of 240 to 700 nm with different wavelength intervals. The lowest detection limits of adulteration, 3.0 and 4.4%, respectively, were observed for the application of wavelength intervals of 60 and 80 nm. Multiple linear regression models were used to calculate the level of adulteration, with the lowest root mean square error of prediction and root mean square error of cross validation equalling 1.5 and 1.8%, respectively, for the measurement acquired at the wavelength interval of 60 nm. Lower classification errors were obtained for the successive projections algorithm-linear discriminant analysis (SPA-LDA) rather than for the principal component analysis (PCA)-LDA method. The lowest classification error rates equalled 3.8% (∆λ = 10 and 30 nm) and 0.0% (∆λ = 60 nm) for the PCA-LDA and SPA-LDA classification methods, respectively. The applied technique is useful for detecting the addition of plant fat to hard cheese.