Portable NIR spectrometer as quick tool for authentication and prediction of the composition of a Brazilian regional butter Manteiga de Garrafa

• Grant number: 20/09653-0
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): November 01, 2020
• Effective date (End): October 31, 2021
• Field of knowledge: Agronomical Sciences - Food Science and Technology - Food Science
• Principal researcher: Douglas Fernandes Barbin
• Grantee: João Pedro Hebling e Tavares
• Home Institution: Faculdade de Engenharia de Alimentos (FEA). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

The Brazilian regional butter "Manteiga de Garrafa" is a product obtained by milk fat, largely appreciated in the North and Northeast of Brazil and in the North/Northeast of Minas Gerais (Brazilian state). Despite its importance for the local economy, it is difficult to find in the literature research on the product. Furthermore, the lack of uniformity in the processes results in adulterations and noncompliance with legislation. This project proposes the use of a portable near-infrared spectrometer (NIR) to evaluate authenticity and predict compositional attributes of "Manteiga de Garrafa". The butter samples will be analyzed to determine the acidity content, peroxide value, moisture, fat content and solids-not-fat content, and spectra will be acquired using a portable NIR spectrometer. The Principal Component Analysis (PCA) will be applied as an exploratory analysis of the data and the discriminant analysis by the method of partial least squares (PLS-DA) will be used to discriminate between authentic and adulterated samples. Through the selection of wavelengths, regression models (PLSR) will be constructed to quantify the presence of adulterants in binary and ternary mixtures of butter with soybean oil, margarine and pork fat. The expected results of this project are the development of techniques to evaluate the composition and authenticity of "Manteiga de Garrafa" by non-invasive methods using multivariate analysis applied to the data obtained from the NIR spectra.