Study and development of smart devices (electronic nose and tongues) aiming at discrimination of contamination in food samples

Author(s)

Lígia Bueno

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Conjunto das Químicas (IQ e FCF) (CQ/DBDCQ)
• Defense date: 2016-04-05

Examining board members

Thiago Regis Longo Cesar da Paixão; Mauro Bertotti; Wendell Karlos Tomazelli Coltro; Fernando Josepetti Fonseca; Nelson Ramos Stradiotto

• Advisor: Thiago Regis Longo Cesar da Paixão

Abstract

The present thesis aimed at development of low cost smart devices (electronic tongue and colorimetric nose) to discriminate chemically and biologically contamination in food samples. An \"optoelectronic\" plastic-based device with colored membranes contained pH indicator was used to discriminate volatile compounds released by microorganisms, due to the deterioration process of protein in food by the organisms. The amines evaluated in this study were: isobutylamine, isopentylamine and triethylamine, achieving a detection limit of 5 ppm. Such system was also tested in real meat samples contaminated with individual amines obtained a good discrimination of samples with and without studied compounds. Biogenic amines (cadaverine, tyramine and putrescine) were also tested and discriminated. In a second step, the device was also evaluated to discriminate four bacteria species (Klebsiella pneumoniae, Proteus vulgaris, Proteus mirabilis and Escherichia coli) incubated at 37 ° C and 25 ° C. In both cases, a smartphone was used as detector to extract RGB values of the samples. From extracted information (RGB values), the chemometric tools PCA (Principal Component Analysis) and HCA (Hierarchical Cluster Analysis) were used to discriminate samples and k-NN (kth Nearest Neighbor) was evaluated to validate the method. In a third stage, a voltammetric electronic tongue was developed to discriminate adulterated milk samples with melamine, urea and formaldehyde. This voltammetric electronic tongue was fabricated using three working electrodes: platinum, gold and copper and the voltammetric data was used as input data for chemometric tools (PCA and HCA). Three types of milk (whole, skimmed and semi-skimmed) from three different brands were tested and all of them could be successfully discriminated (AU)

• FAPESP's process: 11/23355-3 - Study and development of intelligent devices (electronic noses and tongues) aiming to the discrimination of contamination in foods
• Grantee: Lígia Bueno Bedatty
• Support Opportunities: Scholarships in Brazil - Doctorate