Discriminação de farinhas de trigo brasileiras e argentinas empregando espectroscopia e métodos quimiométricos

Abstract

Considered an essential food in the human diet, wheat flour is influenced by geographic origin, grain quality, and the manufacturing process, variables that affect its quality and can be assessed using chemometrics. By employing spectroscopic techniques combined with Principal Component Analysis (PCA), a multivariate chemometric technique, it is possible to identify the main variables that contribute to the variability in flour composition, as well as to identify patterns in the obtained data, quickly and reliably. The main objective of this work was to discriminate commercial wheat flours of Brazilian and Argentine origin, using spectroscopic techniques with reduced sample preparation, such as Fourier Transform Infrared Spectroscopy - Attenuated Total Reflectance (FTIR-ATR), Laser Induced Breakdown Spectroscopy (LIBS), and Energy Dispersive X-ray Fluorescence Spectroscopy (EDXRF), combined with chemometrics. For the FTIR-ATR technique, PCA showed that it is possible to separate the flours based on the country of origin, without any prior treatment, particularly in the spectral region below 1150 cm⁻¹. The Argentine samples exhibited less data dispersion compared to the Brazilian samples and corroborated the results obtained in the moisture analysis. Additionally, partial discrimination of the wheat flour samples based on the cultivar of the wheat grains used was possible. As an additional study, the discrimination between wheat flour and cassava flour at four concentration levels (0%, 3%, 10%, and 40% m/m) was evaluated using the FTIR-ATR technique. For the LIBS and EDXRF analyses, the presence of inorganic elements such as potassium, calcium, and iron were observed, and their importance in the discrimination of the flours by PCA was highlighted, with data variance accumulated exceeding 84%. These results suggest a possible correlation with the location of grain cultivation.