Data fusion of fluorescence and UV spectroscopies improves the detection of cocoa butter adulteration

Abstract

The fraudulent addition of cheaper plant oils to cocoa butter is a real issue for the chocolate industry. The potential of fluorescence and UV and data fusion of these spectroscopies for the detection of cocoa butter adulteration with cocoa butter equivalents (CBEs) is investigated here. Principal component regression (PCR) models have been used to calculate the level of adulteration. A classification model has been built with the help of the principal component analysis (PCA) and the linear discriminant analysis (LDA). The LDA analysis has been applied to fluorescence intensities and UV spectra of cocoa butter, cocoa butter equivalents, and their mixtures in order to examine classification ability. The best prediction ability of adulteration was obtained for PCR data fusion models with the lowest errors of calibration (RMSEC) and validation (RMSEV) of 3.7 and 4.7%. The best performance of the PCA‐LDA analysis has been observed for data fusion of fluorescence and UV intensities measurements at wavelength intervals of 10 and 30 nm. The LDA shows that data fusion can achieve 95.2% correct classifications (sensitivity) in the validation data set, while the corresponding results for individual spectroscopies range from 66.7–90.5%. The results demonstrate that fluorescence and UV spectroscopies complement each other, giving a synergistic effect for the detection of cocoa butter adulteration with cocoa butter equivalents. Practical applications: The aim of this research is to develop a quick and cheap method of discriminating between cocoa butter and cocoa butter equivalents. The obtained results are satisfactory for classification of genuine cocoa butter and cocoa butter equivalents and for detection of cocoa butter adulterations with cheaper cocoa butter equivalents. The method is an important technical support of public health against profit‐driven practices. Application of fluorescence and UV spectroscopies to the detection of cocoa butter adulteration with cocoa butter equivalent and the synergy between both spectroscopies are investigated. The best prediction ability of MLR models i obtained for the data fusion model with the errors of calibration and validation of 3.7 and 4.7%. Compared to individual spectroscopies, data fusion shows better discrimination ability of PCA‐LDA model with the highest classification accuracy equal 100.0 and 95.2% in the calibration and validation data sets. The results show that fluorescence and UV spectroscopies give a synergistic effect for the detection of cocoa butter adulteration.

Authentication of Cocoa Products Based on Profiling and Fingerprinting Approaches: Assessment of Geographical, Varietal, Agricultural and Processing Features

“…Among the arsenal of instrumental techniques available, infrared spectroscopy (IR) and mass spectrometry (MS) are the preferred ones ( Figure 1 a). Other detection techniques, such as nuclear magnetic resonance (NMR), spectroscopies (UV and fluorescence), electroanalytical analyses, or electronic noses, were less exploited in this field [ 20 , 21 , 22 , 23 , 24 , 25 ].…”

Challenges and Opportunities of Implementing Data Fusion in Process Analytical Technology—A Review

“… [ 187 ] Detect adulteration of cocoa butter Fluorescence; UV LLDF PCA-LDA - DF-individual model comparison LLDF > individual model e.d. [ 188 ] Storage time classification Dielectric spectroscopy; Computer Vision MLDF ANN; SVM; BN; MLR CFS; image processing—red, green, blue, hue, saturation, intensity, lightness, a∗ and b∗ chromatic components / MLDF > individual model e.d. [ 189 ] Understand the effect of storage factors on rice germ shelf life NIR; e-nose MLDF PCA PLS (NIR); Pearson’s correlation coefficient-based data selection (e-nose) Correlation maps no comparison / [ 128 ] Characterisation of black pepper LC-MS; GC–MS; NMR MLDF OPLS-DA OPLS-DA -> VIP DF-individual model comparison enhanced process control e.d.…”

Discrimination of Boletaceae mushrooms based on data fusion of FT-IR and ICP–AES combined with SVM

“…For example, Silvestri et al [23] used low-level data fusion strategy for the sake of presenting the geographical variability of wines and obtained a good result. Dankowska [24] gave a synergistic effect for the detection of cocoa butter adulteration with cocoa butter equivalents by using mid-level data fusion of fluorescence and UV spectroscopies. Sun et al [25] established a method by data fusion of NIR and mid-infrared spectra to distinguish unofficial rhubarbs, and the result indicated that the data fusion strategies could improve the classification accuracy.…”