Uso de Espectroscopia Óptica aliada à Inteligência Artificial para identificação e quantificação de adulterantes em polpa de açaí

Abstract

In this work it’s proposed the identification/separation and quantification of adulterants present in açaí pulp samples through spectroscopy techniques associated with chemometric methods. Fluorescence Spectroscopy as well as Laser Induced Breakdown Spectroscopy (LIBS) were used. The samples were contaminated in a controlled way through the monitored addition of three different adulterants (wheat flour, ice cream stabilizer, maize starch) in order to evaluate the efficiency of each of the techniques. For the multidimensional data processing the methods Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used. The best results were obtained through the use of LIBS, associated with PCA and LDA, that was capable to separate the different contaminants from the samples of even the lowest concentration analyzed (2%). This work is part of a scientific collaboration between the Laboratory of Optics and Quantum Information (LOIQ), from the Institute of physics at UFU, and the Laboratory GAIA (Group of Applied Instrumental Analysis) from the department of chemistry at UFSCar (Federal University of São Carlos).