**2.4 Final remarks**

The results described above, showed that PCA conducted over a set of UV absorption spectra from different types of mezcals, allows an efficient and reliable discrimination between artificially and naturally maturated mezcals in wood casks, the data pre-treatments playing an important role for the correct interpretation. This discrimination power is based on the differential absorbance spectra of the compounds naturally produced from wood during maturation stage and those corresponding to the compounds used for adulteration (*i.e.:* colorants to confer a matured appearance to the beverage).

On the other hand, PCA conducted over fluorescence spectra allowed the identification of two main groups correlated with the presence or absence of certain organic molecules, not 30 Principal Component Analysis

Fig. 5 (a) depicts six representative fluorescence spectra corresponding to each type of mezcal analysed. A high similarity between the fluorescence spectrum of ethanol and white/young mezcal is observed. On the other hand, rested, abocado and distilled mezcals, have similar spectra. Finally, the huge differences between the intensity of the emission spectra corresponding to aged mezcal, and that of the other types of mezcal, can be attributed to the higher concentration of organic molecules coming from the wood cask

This also explains the grouping along PC1-axis. From these results, it can be concluded that PC1 can discriminate between naturally/artificially matured samples and white mezcal samples. On the contrary, the one-dimensional PC1-loading plot obtained from standardized spectra; does not provide clear information about the objects distribution.

In this sense, and according with the results described above, the standardization pretreatment does not improve the discrimination between samples in the scores plots. On the

**Abocado**

**560 600 640 680 720 760 800**

**Wavelength (nm)**

Fig. 5. (a) Representative raw fluorescence spectra for each type of mezcals, (b) PC1-loadings

The results described above, showed that PCA conducted over a set of UV absorption spectra from different types of mezcals, allows an efficient and reliable discrimination between artificially and naturally maturated mezcals in wood casks, the data pre-treatments playing an important role for the correct interpretation. This discrimination power is based on the differential absorbance spectra of the compounds naturally produced from wood during maturation stage and those corresponding to the compounds used for adulteration

On the other hand, PCA conducted over fluorescence spectra allowed the identification of two main groups correlated with the presence or absence of certain organic molecules, not

plot for the centred spectra, and (c) PC1-loadings plot for standardized spectra.

(*i.e.:* colorants to confer a matured appearance to the beverage).

*Standardized*

*mean centered*

*raw spectra*

**c**

**b**

**a**

**Ethanol**

contrary, it leads to a misinterpretation of the loading plots.

**Aged**

**DistilledRested**

**White**

**0.014 0.021 0.028 0.035 0.00 0.02 0.04 0.06 0 150 300 450**

**PC1- loadings**

**2.4 Final remarks** 

**PC1- loadings**

**Intensity**

**2.3.2 PCA-loadings** 

during the maturation process.

necessarily correlated with maturation in the wood casks. Thus, fluorescence spectroscopy did not demonstrate to have enough sensibility to discriminate between authentic and artificially matured mezcals.
