Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/56197
Type: Artigo de periódico
Title: Comparison and application of near-infrared (NIR) and mid-infrared (MIR) spectroscopy for determination of quality parameters in soybean samples
Author: Ferreira, DS
Galao, OF
Pallone, JAL
Poppi, RJ
Abstract: Grain composition is directly related to maintenance of quality. Chemical analyses have been determined using traditional and laborious methods, which are time-consuming and generate chemical waste. This justifies the development of fast and accurate alternative methodologies to control the grain composition. Near-infrared (NIR) and mid-infrared (MIR) spectroscopy techniques associated with chemometric tools have been applied in the development of several analytical methodologies for agricultural products. The aim of this study is to develop and compare these two spectroscopic techniques to determine the parameters of quality, such as moisture, protein, lipid and ash content, in 20 varieties of soybean, which are grown in the cities of Ponta Grossa and Londrina, Brazil, totally 40 samples. It was used near-infrared and mid-infrared spectroscopy, with diffuse reflectance measurements, associated with multivariate calibration methods based on partial least squares algorithm. The determination coefficient (R-2) for moisture, ash, protein and lipid content were 0.72, 0.73, 0.88 and 0.81 for NIR and 0.63, 0.87, 0.91 and 0.67 for MIR, respectively, having an RMSECV (root mean square error of cross-validation) <2.09%. The results show that both infrared (NIR and MIR) techniques have predictive abilities. (C) 2013 Elsevier Ltd. All rights reserved.
Subject: Infrared spectroscopy
Soybean
Partial least square
Food quality
Country: Inglaterra
Editor: Elsevier Sci Ltd
Rights: fechado
Identifier DOI: 10.1016/j.foodcont.2013.07.010
Date Issue: 2014
Appears in Collections:Artigos e Materiais de Revistas Científicas - Unicamp

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.