A Sensor-Based Methodology to Differentiate Pure and Mixed White Tequilas Based on Fused Infrared Spectra and Multivariate Data Treatment
Metadata
Show full item recordEditorial
MDPI
Materia
White Tequila Authentication IR spectroscopy Chemometrics Data fusion Beverage alcoholic industry
Date
2021-02-27Referencia bibliográfica
Pérez-Beltrán, C.H.; Zúñiga-Arroyo, V.M.; Andrade, J.M.; Cuadros-Rodríguez, L.; Pérez-Caballero, G.; Jiménez-Carvelo, A.M. A Sensor-Based Methodology to Differentiate Pure and Mixed White Tequilas Based on Fused Infrared Spectra and Multivariate Data Treatment. Chemosensors 2021, 9, 47. [https://doi.org/10.3390/chemosensors9030047]
Sponsorship
Universidad Nacional Autonoma de Mexico PIAPI 2042 PAPIIT IT200918Abstract
Mexican Tequila is one of the most demanded import spirits in Europe. Its fast-raising worldwide request makes counterfeiting a profitable activity affecting both consumers and legal distillers. In this paper, a sensor-based methodology based on a combination of infrared measurements (IR) and multivariate data analysis (MVA) is presented. The case study is about differentiating two categories of white Tequila: pure Tequila (or '100% agave') and mixed Tequila (or simply, Tequila). The IR spectra were treated and fused with a low-level approach. Exploratory data analysis was performed using PCA and partial least squares (PLS), whilst the authentication analyses were carried out with PLS-discriminant analysis (DA) and soft independent modeling for class analogy (SIMCA) models. Results demonstrated that data fusion of IR spectra enhanced the outcomes of the authentication models capable of differentiating pure from mixed Tequilas. In fact, PLS-DA presented the best results which correctly classified all fifteen commercial validation samples. The methodology thus presented is fast, cheap, and of simple application in the Tequila industry.