Teófilo R. F., Cardoso M. G., Vilela F. J., de Lima A. B., Ribeiro C. F. S., Silva V. F., Ferreira M. M. C., "EVALUATION OF THE QUALITY OF DOMESTIC CACHAÇA (SUGAR-CANE SPIRITS) FROM MINAS GERAIS APPLYING PRINCIPAL COMPONENT ANALYSIS". São Carlos, SP, Brazil, 01-04/12/2004: V Brazilian Meeting on Chemistry of Food and Beverages (V BMCFB), Book of Abstracts (2004). Poster. Session: Quality control.

EVALUATION OF THE QUALITY OF DOMESTIC CACHAÇA
(SUGAR-CANE SPIRITS) FROM MINAS GERAIS APPLYING PRINCIPAL
COMPONENT ANALYSIS
 

Reinaldo F. Teófilo^b*, Maria das G. Cardoso^a*, Fernando J. Vilela, Annete B. de Lima^a,
Cleuza de F. S. Ribeiro^a, Vanisse de F. Silva^a, Márcia M. C. Ferreira^b

^aDQI, Universidade Federal de Lavras, mcardoso@ufla.br
^bIQ, Universidade Estadual de Campinas, marcia@iqm.unicamp.br
 

        The  increase  in  the  consumpiton  of  sugar  cane  spirits  -  the  cachaça  -  with
good quality  and the possibility of exportation require  that  the  fabrication  processs of
this drink  must  be  based in practices determined  in  a criterion way.   As a result,  the
Ministério da Agricultura, Pecuária  e  Abastecimento  (MAPA)  establishes  limits  to
some   cachaça  parameters  based   on    its  alcoholic   graduation  (AG),   the  volatile
components sum (VCS),  such as  aldehydes (Al),  acids (Ac),  esters (Es),  furfural and
superior  alcohols  (AS),   and  the  maximum  drifts  allowed  to  methyl  alcohol  (Me),
copper ion (Cu²⁺)  and  volatile  individual components.  The  cachaça  does not obey to
the MAPA quality patterns when it does not respect  at  least o ne of the defined  limits¹.
        The aim of this work was  to  evaluate  the  domestic  cachaça  from Minas Gerais
that were  analyzed  by  the MAPA methodology.   To  this  evaluation  were  used  the
Principal  Component  Analysis  (PCA)   as   the   non  supervised   pattern  recognition
method.
 


 

Figure 1. Score and loadings graphics to the factors 1 and 2.

        It is about 94 samples and 8 variables that were analyzed. According  to the matrix
(94x8) built, it was realized the PCA method with the auto-scaled data.   In the figure  1,
the  score  graphics,  the  A  one,  indicates  the  cachaça samples that respected  to  the
required patterns  were situated in the center,  while the other ones,  that did not respect
the patterns specified, were distributed around the graphics.  The B graphic, the  loading
one,  indicated which variables  contributed  to  this  distribution,   only being these ones
the AS, VCS,  Al,  Ac,  and  Cu²⁺  that presented influence to the  discrimination among
the cachaças.  Certainly  to  this  fact,  they  were  the most  important variables  to  the
quality  evaluation.   The  variables  AG,  Es   and   Me   were  not  influencial    to  the
discrimination and because  of  this  they presented similar values,  generally  respecting
the  defined  limits.
______________________________________________________________________
  The authors would like to thank to the CNPq for the financial support.
______________________________________________________________________
1. Cardoso, M. G. Produção de Aguardente de Cana-de-acúcar, Editora UFLA: Lavras,
2001; pp 152-167.
2. Beebe, K. R.; Pell, R. J.; Seasholtz, M. B. Pattern Recognition. In Chemometrics: A
Practical Guide; Wiley: New York, 1998; pp 62-125.