Development of 2D and 3D front face fluorescence spectroscopy for monitoring ultrasound treatment in the removal of pesticides residues from fresh lettuces at the laboratory and pilot scales.

Pesticide residues in vegetables are potentially toxic components to humans and can cause serious health problems. To remove pesticide residues from fresh agricultural products and improve consumer food safety, various pesticide removal methods have been investigated over the past decades. In this study, the effectiveness of laboratory and pilot scale ultrasonic cleaning on the removal of boscalid and pyraclostrobin residues from lettuce was examined. 2D fluorescence spectroscopy, 3D fluorescence spectroscopy represented by excitation-emission matrix (EEM), and parallel factor analysis (PARAFAC) were used to characterize and discriminate the fluorescence signatures of these pesticides in the cleaning water to determine the effectiveness of the ultrasonic cleaning method as a function of the level of pesticide removal. The 2D fluorescence results showed that the rate of removal of boscalid by ultrasonics at the laboratory scale increased with the cleaning time. The ultrasonic treatment showed a higher cleaning efficiency compared to only soaking in distilled water for 10 min. The same trends were observed at the pilot scale. The EEM also showed differences in the concentration of pesticides removed by ultrasonication between the different parts of the lettuce, the concentration was higher in the upper part than the lower part. This study showed that ultrasonication is an effective technique for the removal of pesticide residues on lettuce, and it also showed the significant potential of fluorescence spectroscopy coupled with PARAFAC for the discrimination and characterization of pesticides.

Impact of storage period and lipid unsaturation on the kinetic of 5-hydroxymethylfurfural and furfural generation in pound cakes.

This work aimed at studying the influence of formulation and storage on the formation of Maillard compounds called 5-hydroxymethylfurfural (HMF) and furfural in pound cakes formulated with rapeseed oil (RO) and palm oil (PO). A progressive humidification of the crust and a dryness of the crumb were observed during storage. Lightness (L*) decreased for both PO and RO pound cakes in crumb as well as crust. The accumulation of HMF was found to be higher in the crust (2.21-38.5 mg kg-1) in comparison with the crumb (0.78 to 10.29 mg kg-1). Similar results were found for furfural where the concentration range was between 0.98 and 5.67 mg kg-1 in the crumb while between 2.1 and 38.39 mg kg-1 in the crust. Thus, the formation of HMF and furfural depended on formulation. The PLS-DA showed a positive correlation between the kinetic of HMF and furfural migration and the storage period.

Potentiality of front-face fluorescence and mid-infrared spectroscopies coupled with partial least square regression to predict lipid oxidation in pound cakes during storage.

The potentialities of front-face fluorescence (FFF) and mid-infrared (MIR) spectroscopies coupled with partial least square regression (PLSR) were compared to predict the lipid oxidation of pound cakes. The level of lipid oxidation in pound cakes determined using classical methods showed some changes. Similarly, the fluorescence emission (305-490 nm) and excitation (252-390 nm) spectra and MIR spectra scanned in the 4000-700 cm-1 region showed some changes in pound cakes as a function of both storage time and the type of oil used in the formulation. The application of PLSR to the MIR spectra, provided excellent predictive results for free fatty acid (R2 = 0.97) and peroxide values (R2 = 0.87). Similar results were obtained from both tryptophan and MIR spectra for the prediction of TOTOX (R2 > 0.86) demonstrating the efficiency of the MIR and FFF spectroscopies to qualify and quantify the level of lipid oxidation in pound cakes.

Use of front face fluorescence for monitoring lipid oxidation during ageing of cakes.

Front face fluorescence spectroscopy coupled with chemometric tools was used as a useful tool for the monitoring of sponge cakes freshness, produced at the pilot scale, during ageing (i.e. 1, 3, 6, 9, 16, and 20days). The fluorescence emission spectra were acquired in the 340-490nm and 390-680nm after excitation at 325 and 380nm, respectively, while excitation spectra (250-390nm) were scanned after emission at 410nm. The primary and secondary products of lipid oxidation were also determined on the same cakes. The principal component analysis (PCA) applied to the each spectral collection obtained after excitation at 325 and 380nm and emission at 410nm allowed a clear discrimination of cakes according to their ageing. A high correlation between the intensity of fluorescence at 521nm and the p-anisidine values was observed since squared correlation coefficient of 0.73 was obtained. The results showed that fluorescence spectroscopy could be used as a powerful tool for the evaluation of cake freshness throughout storage.

Monitoring changes in sponge cakes during aging by front face fluorescence spectroscopy and instrumental techniques.

In the present study, sponge cakes, produced at the pilot scale, were monitored during aging (i.e., 1, 3, 6, 9, 16, and 20 days) by three different analytical techniques. For the texture analyzer, the hardness and elasticity of crumb cakes were found to significantly increase and decrease, respectively, throughout aging. Color parameters (L*, a*, and b*) showed only slight change throughout aging, and a high correlation (R(2) = 0.88) was observed between the whiteness and the yellowness. Tryptophan fluorescence spectra (excitation, 290 nm; emission, 305-490 nm) recorded on cakes exhibited three maxima located at 382, 435, and 467 nm that were attributed to maximum emission of tryptophan (382 nm) and fluorescent Maillard reaction products (435 and 467 nm). The principal component analysis (PCA) applied to the tryptophan spectra allowed a clear discrimination of cakes aged for 1, 3, and 6 days from those aged for 9, 16, and 20 days. Finally, canonical correlation analysis (CCA) performed on the textural and tryptophan fluorescence spectral data sets showed that the two groups of variables were highly correlated because the squared canonical coefficients for canonical variates were 0.99, indicating that cake texture determined at the macroscopic level by texture analyzer is a reflection of its structure at the molecular level determined by fluorescence spectroscopy.