RESUMEN

The authentication of Slovak wines in comparison to other similar wines from various geographical regions, namely Hungary, France, Austria, and Ukraine, was conducted using the OC-PLS, DD-SIMCA, and PLS-DM models, all of them operating in rigorous way. The study involved 63 samples, of which 41 originated from Slovakia, covering diverse wine types such as varietal wines, cuvée selections (different "putnový"), and essence. To capture digital images under controlled conditions, a custom-made cardboard box with white inner surfaces was devised and equipped with a smartphone. During the training phase, sensitivities of 96%, 100%, and 96% were attained for OC-PLS, DD-SIMCA, and PLS-DM, respectively. In the subsequent stages of validation and testing for DD-SIMCA and PLS-DM, the proposed methods displayed optimal efficiency, achieving both sensitivity and specificity rates of 100%. However, such results were not achieved in the case of OC-PLS, which exhibited efficiency levels of 90% in validation and 80% in testing.

Asunto(s)

RESUMEN

This study integrates genetic algorithm (GA) with partial least squares regression (PLSR) and various variable selection methods to identify impactful regions of interest (ROI) in heterogeneous 2D chromatogram images for predicting wine age. As wine quality and aroma evolve over time, transitioning from youthful fruitiness to mature, complex flavors, which leads to alterations in the composition of essential aroma-contributing compounds. Chromatograms are segmented into subimages, and the GA-PLSR algorithm optimizes combinations based on grayscale, red-green-blue (RGB), and hue-saturation-value (HSV) histograms. The selected subimage histograms are further refined through interval selection, highlighting the compounds with the most significant influence on wine aging. Experimental validation involving 38 wine samples demonstrates the effectiveness of this approach. Cross-validation reduces the PLS model error from 2.8 to 2.4 years within a 10 × 10 subset, and during prediction, the error decreases from 2.5 to 2.3 years. The study presents a novel approach utilizing the selection of ROI for efficient processing of 2D chromatograms focusing on predicting wine age.

RESUMEN

In this study, we report the simultaneous determination of bromine and fluorine using Second-Order Calibration High-Resolution Continuum Source Graphite Furnace Molecular Absorption Spectrometry (HR CS MAS). The instrumental data acquired correspond to the time versus wavelength matrix per sample that were analyzed using Parallel Factor Analysis (PARAFAC), along with Unfold and N-way Partial Least Squares combined with a post-calibration step known as Residual Bilinearization (U and N PLS/RBL). Despite the significant difference in sensitivity between bromine and fluorine, all approaches provided reasonably accurate results when predicting both analytes in synthetic mixtures within a controlled environment. The relative prediction error (REP) values for bromine were 29.8 % (PARAFAC), 23.6 % (N-PLS/RBL), and 13.1 % (U-PLS/RBL), while for fluorine, the REP values were 3.4 % (PARAFAC), 3.5 % (N-PLS/RBL), and 3.2 % (U-PLS/RBL). When applying this approach to predict unknown samples, a comparison was made between the estimated nominal concentrations of fluorine and bromine obtained using either a reference method or based on labeled values or spiked mass, and those obtained using the proposed method. It was observed that PARAFAC was unable to predict the samples accurately, whereas the REP values for the prediction of bromine and fluorine using N-PLS/RBL and U-PLS/RBL methods were 19.3 %/19.2 % and 13.6 %/13.1 %, respectively.

RESUMEN

One of the most effective methods for gaining insight into the composition of trace-level volatile organic characteristics of wine products is through the use of a comprehensive two-dimensional gas chromatography-high resolution mass spectrometry (GC × GC-HRMS) technique. The vast amount of data generated by this method, however, can often be overwhelming requiring exhaustive and time-consuming analysis to identify significant statistical characteristics. The use of advanced chemometric software can achieve the same or even higher efficiency. This study aimed to identify differences based on geographical locations by analyzing the volatile organic compounds in the composition of botrytized wines from Slovakia, Hungary, France, and Austria. The volatile organic compounds were extracted by solid-phase microextraction and analyzed using GC × GC-HRMS. The data obtained from the analysis underwent Fisher-ratio (F-ratio) tile-based analysis to identify statistically significant differences. Principal component analysis demonstrated a significant distinction between wine samples based on geographical location, using only 10 statistically significant features with the highest F-ratio. In the samples, the following compounds were analyzed: methyl-octadecanoate, 2-cyanophenyl-ß-phenylpropionate, α-ionone, n-octanoic acid, 1,2-dihydro-1,1,6-trimethyl-naphthalene, methyl-hexadecanoate, ethyl-pentadecanoate, ethyl-decanoate, and γ-nonalactone. These, all play an important role in cluster pattern observed on principal component analysis results. Additionally, hierarchical cluster analysis confirmed this.

RESUMEN

In this study, a new approach was developed for classifying grape juices produced in Brazil using unfolded excitation-emission matrix (EEM) fluorescence spectroscopy and chemometrics, with respect to the agricultural production system, namely the conventional or organic agricultural one. Linear discriminant analysis (LDA) coupled to ant colony optimisation (ACO) and the genetic algorithm (GA) were used to select a more effective subset of variables to discriminate grape juice samples. The best results demonstrated highly efficient classification of grape juice samples according to a conventional or organic production process with an accuracy rate of up to 97% for the models and 94% in the prediction of these classes for samples external to the model. The models showed high selectivity and sensitivity with a rate of up to 100% for the training and test datasets, in addition to determining the most significant variables that explain the separation of classes. The proposed method proves to be viable, as it is fast and requires minimal sample preparation, allowing quality control in the food industry.

Asunto(s)

RESUMEN

Multi-stir bar sorptive extraction (multi-SBSE) represents a viable alternative for recent trends in sample preparation based on a combination of extraction techniques. In this case, increased amount of sorbent and its extended polarity range could advance quality of experimental data obtained in foodomics or metabolomics investigations. With this in mind, it was developed multi-SBSE procedure suitable for authentication of botrytized wine produced in different countries of Tokaj wine region. A design of headspace mode of multi-SBSE was modified to provide additional agitation of a stir bar. An expanded profile of wine samples was obtained with the application of EG-Silicone and PDMS coated stir bars in headspace and direction immersion mode, respectively. Multivariate optimization based on central composite design was selected to determine the influence of various experimental parameters, including extraction temperature enhancing headspace extraction. In our case, proper description of the optimization results required application of a third-order polynomial model, which highlighted dominant influence of salt addition for extraction in both modes. Due to a large number of extracted compounds, comprehensive two-dimensional gas chromatography (GC × GC) was used for assessment of wine samples. Such approach allows reveleaing the presence of sulphur containing compounds, diols, ketone derivatives and methoxybenzenes linking a specific geographical origin. At the same time, the results obtained for compounds common for all the samples were processed with principal component analysis (PCA). Considerable progress for discrimination of the botrytized wines was mainly achieved with combined data from EG-Silicone and PDMS extraction.

Asunto(s)

RESUMEN

In spite of extensive applications of flow modulated comprehensive two-dimensional gas chromatography (FM-GG × GC) in different research areas, its application in the field of chiral separation is very limited. From a practical point of view, the establishment of experimental parameters for enantiomer separations is possibly more demanding in this case. Since the carrier gas flows in both dimensions, it affects not only the separation parameters, but also the fill/flush volumes of the modulator and its working efficiency. In this context, a multivariate design of experiment was applied to find the optimum experimental parameters of a reversed fill/flush (RFF) modulator for enantiomer separation of organic compounds present in botrytized wine samples. The results were described both with response surface methodology and artificial neural networks (ANN). The enantiomeric composition of chiral compounds present in the botrytized wines was used to identify their geographical origin, by principal component analysis (PCA). In addition, the developed one-class partial least squares (OC-PLS) model enabled recognition of the wine samples from the Tokaj wine region with 93% effectiveness in the presence of other samples.

Asunto(s)

RESUMEN

New approach to deal with food authentication by modelling methods based on data recorded from different sources is proposed and called OC-PLS, combines an orthogonalization step between the different data sets to eliminate redundant information followed by definition of an acceptance area for a target class by OC-PLS. The proposed method was evaluated in two case studies. The first study used a controlled scenario with simulated data. In the second case study, the approach was applied using UV-VIS and IR data, in order to differentiate Slovak Tokaj Selection wines of high quality from other lower market value wines from the Slovak Tokaj wine region. In both cases, better results were reached than when individual blocks of data were achieved. The proposed method proved to be effective in properly exploring common and distinct information in each data block. The best compromise between sensitivity and selectivity in the prediction step was achieved.

Asunto(s)

RESUMEN

The main critical point of newly developed miniaturized sample preparation techniques is a limited extraction capacity. Dynamic headspace extraction offers increased volume of sorbent which is commonly used in environmental analysis. Application of two sorbents (Carbopack B/Carbopack X and Tenax® TA) at different extraction temperatures allows enhancing a range of volatile organic compounds available for analysis. Such approach was applied in our research for quantification of volatile organic compounds in botrytized wines with gas chromatography. The central composite design was included to analysis simultaneous effects of incubation time, incubation temperature, purge volume and purge flow. In attempt to properly assess results, the data evaluation involved Pareto charts, surface response methodology and principal component analysis. Multivariate experimental design revealed statistical significance of purge volume and quadratic terms of incubation time and temperature, for response of volatiles. The quantification method with 0.2-2.0 µg/L LOD and 0.5-5.0 µg/L LOQ values, was developed under simultaneously optimized experimental conditions such as a 54 °C incubation temperature, a 20.18 min incubation time, a 344.3 mL purge volume and a 16.0 mL/min purge flow. The increased levels of linalool oxide, ethyl phenylacetate, γ-hexalactone and α-terpineol were observed in the samples, that correlated with botrytized wine technology.

Asunto(s)

RESUMEN

Tokaj wines could be produced only in so called Tokaj/Tokay wine region that falls within two countries Slovakia and Hungary. Thus, wines bearing Tokaj appellation must be produced only in Hungary and Slovakia by traditional process. Unfortunately, some counterfeit wines from neighbour region in Ukraine could be found in market. The aim of this work is to explore a simple UV-VIS spectrum to recognise true Tokaj/Tokay wines from counterfeits and try to differentiate wines based on their country of origin. This type of question can be duly answered using one class classification approach. Two different approaches, Data Driven Soft Independent Modelling of Class Analogy - DD-SIMCA and One-Class Partial Least Squares - OC-PLS were tested and evaluated for this purpose. In both cases, rigorous way models were built and optimized using only samples of the target class. A set of external samples containing samples from target class and non-target were used to validate the models ability to recognize Slovak samples and reject non-Slovak samples. Model based on DD-SIMCA showed better performance (97% correct rating) compared to OC-PLS models (80% correct rating). Comparing both approaches in terms of sensitivity and specificity, both exhibit high sensitivity (low false negative rate: DD-SIMCA 95% and OC-PLS 100%), however the OC-PLS based model showed low specificity (40%) while DD-SIMCA showed high specificity (100%) rejecting all samples out of Slovak origin. Therefore, the results found in this study show that it is possible to successfully combine UV-VIS spectra and DD-SIMCA models to discriminate Tokaj wine samples of Slovak origin from others. Equally important is environmentally friendly (fast, simple, absence of solvents) classification method in line with green chemistry.

Asunto(s)

RESUMEN

The Tokaj Selection wines (hungarian equivalent "Aszú") are typical noble sweet wines produced in Tokaj wine region that falls within two countries, Slovakia and Hungary. Taking into account the economic importance and uniqueness of these wines, in this work, a new, fast and inexpensive method that combines infrared spectroscopy and multivariate models for characterization Slovak Tokaj Selection wines was developed. At first, sample authentication via one class models (dd-SIMCA) considered Slovak Tokaj Selection wines as target class. The non-target sample was considered to be only a Tokaj sample of Slovak origin. The resulting model was able to properly recognize samples of the target class with high sensitivity and specificity. Subsequently, the putna index was predicted via PLS models. RMSEP equals 0.44; REPpred of 9.6 and R2 0.95 was found in prediction step.

RESUMEN

This paper describes, by the first time, a chemometric approach that combines a simple set of the UV-Vis spectra and partial least square regression (PLSR) for measuring the removal of five pharmaceuticals present in simulated hospital effluents by sorption using activated carbon. The use of multivariate calibration allowed the quantification of the remaining concentrations of the studied drugs present in a complex mixture with high accuracy, avoiding the need for the use of sophisticated methodologies based on chromatography. Isothermal sorption studies were performed on single-component solutions containing amoxicillin, paracetamol, propranolol, sodium diclofenac, or tetracycline as well as on a solution containing a mixture of all these 5 compounds. The isotherm data obtained were fitted to the Langmuir, Freundlich and Liu models. It was observed that for each pharmaceutical, the maximum sorption capacity of the activated carbon was higher for the single component than in the mixture. It was observed that the removal of paracetamol, propranolol, and tetracycline, the removal was complete (100%) and for amoxicillin and sodium diclofenac it was at least 92.71 ±â€¯3.15% and 91.82 ±â€¯0.95% respectively, indicating that the avocado seed activated carbon is an adsorbent with high sorption capacity that can remove five pharmaceuticals from simulated hospital effluents.

RESUMEN

This work proposes an analytical strategy utilizing digital images (DI) for the iron inorganic speciation in white wine. The method was established by the reaction of iron(II) ions with 1,2 ortho-phenanthroline as a chromogenic reagent. Total iron was determined using the same reagent after the addition of hydroxyl ammonium chloride as a reducing agent. In both cases, digital images of the standards/chromogenic reagent and samples were acquired and stored in JPEG format. The region of interest (ROI) was determined with a constant square shape for all images. The ROI was submitted to decomposition in color values according to the RGB additive color model. However, the data obtained by the blue channel was the one used in the construction of the analytical curves because it presented the highest sensitivity. The optimization of the experimental conditions of the procedure was performed by employing multivariate techniques. The precision was evaluated using a wine sample with iron (II) and total iron contents of 0.41 and 0.69 mg L-1, respectively. The results expressed as relative standard deviations were 3.57% for iron (II) and 4.76% for total iron contents. A comparison between the results obtained for total iron by the DI method with the results found using flame atomic absorption spectrometry confirmed the method accuracy. The DI procedure was applied for speciation analysis in six white wine samples and the contents found varied from 0.41 to 1.67 mg L-1 for iron (II) and from 0.69 to 1.71 mg L-1 for total iron. These results are in agreement with those found for speciation analysis of iron in wine samples. Iron (III) contents can be found by the difference between the total iron and iron (II) contents.

RESUMEN

A rapid and non-destructive methodology is proposed for the screening of edible vegetable oils according to conservation state expiration date employing near infrared (NIR) spectroscopy and chemometric tools. A total of fifty samples of soybean vegetable oil, of different brands andlots, were used in this study; these included thirty expired and twenty non-expired samples. The oil oxidation was measured by peroxide index. NIR spectra were employed in raw form and preprocessed by offset baseline correction and Savitzky-Golay derivative procedure, followed by PCA exploratory analysis, which showed that NIR spectra would be suitable for the classification task of soybean oil samples. The classification models were based in SPA-LDA (Linear Discriminant Analysis coupled with Successive Projection Algorithm) and PLS-DA (Discriminant Analysis by Partial Least Squares). The set of samples (50) was partitioned into two groups of training (35 samples: 15 non-expired and 20 expired) and test samples (15 samples 5 non-expired and 10 expired) using sample-selection approaches: (i) Kennard-Stone, (ii) Duplex, and (iii) Random, in order to evaluate the robustness of the models. The obtained results for the independent test set (in terms of correct classification rate) were 96% and 98% for SPA-LDA and PLS-DA, respectively, indicating that the NIR spectra can be used as an alternative to evaluate the degree of oxidation of soybean oil samples.

Asunto(s)

RESUMEN

This paper investigates the use of UV-vis, near infrared (NIR) and synchronous fluorescence (SF) spectrometries coupled with multivariate classification methods to discriminate biodiesel samples with respect to the base oil employed in their production. More specifically, the present work extends previous studies by investigating the discrimination of corn-based biodiesel from two other biodiesel types (sunflower and soybean). Two classification methods are compared, namely full-spectrum SIMCA (soft independent modelling of class analogies) and SPA-LDA (linear discriminant analysis with variables selected by the successive projections algorithm). Regardless of the spectrometric technique employed, full-spectrum SIMCA did not provide an appropriate discrimination of the three biodiesel types. In contrast, all samples were correctly classified on the basis of a reduced number of wavelengths selected by SPA-LDA. It can be concluded that UV-vis, NIR and SF spectrometries can be successfully employed to discriminate corn-based biodiesel from the two other biodiesel types, but wavelength selection by SPA-LDA is key to the proper separation of the classes.

Asunto(s)

RESUMEN

This work proposes a method for monitoring the ageing of beer using near-infrared (NIR) spectroscopy and chemometrics classification tools. For this purpose, the Successive Projections Algorithm (SPA) is used to select spectral variables for construction of Linear Discriminant Analysis (LDA) classification models. A total of 83 alcoholic and non-alcoholic beer samples packaged in bottles and cans were examined. To simulate a long storage period, some of the samples were stored in an oven at 40°C, in the dark, during intervals of 10 and 20 days. The NIR spectrum of these samples in the range 12,500-5405 cm(-1) was then compared against those of the fresh samples. The results of a Principal Component Analysis (PCA) indicated that the alcoholic beer samples could be clearly discriminated with respect to ageing stage (fresh, 10-day or 20-day forced ageing). However, such discrimination was not apparent for the non-alcoholic samples. These findings were corroborated by a classification study using Soft Independent Modelling of Class Analogy (SIMCA). In contrast, the use of SPA-LDA provided good results for both types of beer (only one misclassified sample) by using a single wavenumber in each case, namely 5550 cm(-1) for non-alcoholic samples and 7228 cm(-1) for alcoholic samples.

Asunto(s)

RESUMEN

This work is concerned of evaluate the use of visible and near-infrared (NIR) range, separately and combined, to determine the biodiesel content in biodiesel/diesel blends using Multiple Linear Regression (MLR) and variable selection by Successive Projections Algorithm (SPA). Full spectrum models employing Partial Least Squares (PLS) and variables selection by Stepwise (SW) regression coupled with Multiple Linear Regression (MLR) and PLS models also with variable selection by Jack-Knife (Jk) were compared the proposed methodology. Several preprocessing were evaluated, being chosen derivative Savitzky-Golay with second-order polynomial and 17-point window for NIR and visible-NIR range, with offset correction. A total of 100 blends with biodiesel content between 5 and 50% (v/v) prepared starting from ten sample of biodiesel. In the NIR and visible region the best model was the SPA-MLR using only two and eight wavelengths with RMSEP of 0.6439% (v/v) and 0.5741 respectively, while in the visible-NIR region the best model was the SW-MLR using five wavelengths and RMSEP of 0.9533% (v/v). Results indicate that both spectral ranges evaluated showed potential for developing a rapid and nondestructive method to quantify biodiesel in blends with mineral diesel. Finally, one can still mention that the improvement in terms of prediction error obtained with the procedure for variables selection was significant.

Asunto(s)