TY - JOUR
T1 - Feasibility of different chemometric techniques to differentiate commercial Brazilian sugarcane spirits based on chemical markers
AU - Granato, Daniel
AU - de Oliveira, Camila Cardoso
AU - Caruso, Miriam Solange Fernandes
AU - Nagato, Letícia Araújo Farah
AU - Alaburda, Janete
PY - 2014/6
Y1 - 2014/6
N2 - Chemometric techniques were used to assess the quality of 51 commercial Brazilian sugarcane spirits (cachaça) based on chemical markers. Benzo(a)pyrene, methanol, 2,3-methyl-1-butanol, acetaldehyde, isobutyl alcohol, n-propanol, density, alcoholic strength, and higher alcohols were quantified using chromatographic methods and results were subjected to Principal Component Analysis (PCA), Hierarchical Cluster Analysis (HCA), and Linear Discriminant Analysis (LDA). No significant differences (p. ≥. 0.071) were observed in the chemical markers of Brazilian aged and non-aged cachaça samples. Besides non-significant (p = 0.922), the content of benzo(a)pyrene in aged sugarcane spirits was 1.83 times higher than in non-aged ones. Differences in alcoholic strength (p = 0.001) and n-propanol (p = 0.015) were observed among cachaças produced by double distilling, alembic and in stainless steel columns. PCA was not suitable to separate the samples according to the provenance, aging and distilling process, while HCA was effective in separating alembic cachaças produced by from two distinct producing regions. LDA seemed to be very suitable to assess not only the provenance but also the distilling and aging processes that cachaça undergoes, yielding about 91% accuracy to discriminate non-aged from aged cachaça, 81.82% and 86.61% accuracy to discriminate samples from Minas Gerais and São Paulo, respectively.
AB - Chemometric techniques were used to assess the quality of 51 commercial Brazilian sugarcane spirits (cachaça) based on chemical markers. Benzo(a)pyrene, methanol, 2,3-methyl-1-butanol, acetaldehyde, isobutyl alcohol, n-propanol, density, alcoholic strength, and higher alcohols were quantified using chromatographic methods and results were subjected to Principal Component Analysis (PCA), Hierarchical Cluster Analysis (HCA), and Linear Discriminant Analysis (LDA). No significant differences (p. ≥. 0.071) were observed in the chemical markers of Brazilian aged and non-aged cachaça samples. Besides non-significant (p = 0.922), the content of benzo(a)pyrene in aged sugarcane spirits was 1.83 times higher than in non-aged ones. Differences in alcoholic strength (p = 0.001) and n-propanol (p = 0.015) were observed among cachaças produced by double distilling, alembic and in stainless steel columns. PCA was not suitable to separate the samples according to the provenance, aging and distilling process, while HCA was effective in separating alembic cachaças produced by from two distinct producing regions. LDA seemed to be very suitable to assess not only the provenance but also the distilling and aging processes that cachaça undergoes, yielding about 91% accuracy to discriminate non-aged from aged cachaça, 81.82% and 86.61% accuracy to discriminate samples from Minas Gerais and São Paulo, respectively.
KW - Cachaça
KW - Contaminants
KW - Hierarchical cluster analysis
KW - Linear discriminant analysis
KW - Principal component analysis
UR - http://www.scopus.com/inward/record.url?scp=84888062267&partnerID=8YFLogxK
U2 - 10.1016/j.foodres.2013.09.044
DO - 10.1016/j.foodres.2013.09.044
M3 - Article
AN - SCOPUS:84888062267
SN - 0963-9969
VL - 60
SP - 212
EP - 217
JO - Food Research International
JF - Food Research International
ER -