Characterization of the aromatic profile of the Quebranta variety of Peruvian pisco by gas chromatography-olfactometry and chemical analysis

Method development of stir bar sportive extraction coupled with thermal desorption-gas chromatography-mass spectrometry for the analysis of phthalates in Peruvian pisco

In this work, for the first time, a stir bar sorptive extraction (SBSE) coupled with thermal desorption-gas chromatography-mass spectrometry (TD-GC-MS) was developed and validated for the determination of seven phthalates in Peruvian pisco. The phthalate compounds considered were dimethyl phthalate (DMP), diethyl phthalate (DEP), bis(2-ethylhexyl) hexahydrophthalate (BEHP), benzyl butyl phthalate (BBP), di-n-butyl phthalate (DBP), di-isodecyl phthalate (DIDP) and di-isobutyl phthalate (DIBP). The best overall analytical conditions obtained from the optimization were as follow: extraction time of 120 min, size of polydimethylsiloxane (PDMS) twister (20 mm length x 1 mm thickness), NaCl content (20 %) and sample volume (40 mL). The in-house validation of SBSE/TD-GC-MS method was performed taking into account the ISO/IEC 17,025 requirements and EURACHEM/CITAC guideline. Under optimal conditions, very low limits of detection of 1.3-0.21 µg L-1 were obtained. Furthermore, the limits of quantification ranged from 4.2-70 µg L-1, and the correlation coefficients were found to be ≥ 0.991. The method was precise, with relative standard deviations (RSD, %) for inter twister repeatability and the inter day repeatability precisions from 1.1 to 11 and from 6.2 to 15.9, respectively. The pisco samples were analysed with recoveries between 91-124.4%, and DBP, BEHP, and BBP were the most commonly found compounds in the samples. The optimized methodology was also evaluated in terms of green character, and it obtained almost the best AGREE score when it was compared with other previous methods for the analysis of phthalates in alcoholic beverages. Therefore, the SBSE/TD-GC-MS method has proved to be suitable for routine practice because it is simple, less laborious, economical, precise, accurate and green, and it would be applicable for pisco safety regulations.

Application of Machine Learning Algorithms to Classify Peruvian Pisco Varieties Using an Electronic Nose

Pisco is an alcoholic beverage obtained from grape juice distillation. Considered the flagship drink of Peru, it is produced following strict and specific quality standards. In this work, sensing results for volatile compounds in pisco, obtained with an electronic nose, were analyzed through the application of machine learning algorithms for the differentiation of pisco varieties. This differentiation aids in verifying beverage quality, considering the parameters established in its Designation of Origin". For signal processing, neural networks, multiclass support vector machines and random forest machine learning algorithms were implemented in MATLAB. In addition, data augmentation was performed using a proposed procedure based on interpolation-extrapolation. All algorithms trained with augmented data showed an increase in performance and more reliable predictions compared to those trained with raw data. From the comparison of these results, it was found that the best performance was achieved with neural networks.