Influence of type of distillation apparatus on chemical profiles of brazilian cachaças.

Brazilian cachaças (115 samples; 73 samples derived from distillation in copper pot stills, 42 samples derived from distillation in stainless steel columns), collected directly at the producers, were analyzed for the contents of 34 constituents by chromatography, inductively coupled plasma optical emission spectrometry, and atomic absorption spectrometry. The analytical data were subjected to principal component analysis (PCA) and linear discriminant analysis (LDA). The PCA treatment led to discrimination of the two groups of cachaças, explaining 65.0% of the database variance. Using LDA and ethanal, ethyl carbamate, dimethyl sulfide, isobutyl alcohol, n-propanal, copper, ethyl acetate, and phenylmethanal as chemical descriptors, a model was developed that presented 95.1% accuracy in predicting the type of distillation apparatus.