Preparation of multivariate calibration models for various types of nuts

Identifier:  TFG:305

Handle: https://hdl.handle.net/20.500.11797/TFG305

Authors:  Ramos Nadal, Josep Lluís

Abstract:
Near Infrared (NIR) reflectance spectroscopy has been established itself as an important analytical technique in the field of food and agriculture. In this case, it is used as a rapid and non-destructive method to determine the content of many parameters in almonds, hazelnuts and their elaborated products. These parameters are moisture, fat content and linoleic acid. Firstly, an introduction about NIR spectroscopy is shown, followed by an explanation of the parameters that will be determined. Also the importance of the determination of these parameters in food industry is mentioned. Next, the reference methods and Partial Least-Squares (PLS) concepts are explained. To determine the moisture in solid samples a Halogen Moisture Analyzer is used and for paste samples an oven at 105ºC is used. Fat content in solid samples is determined by Nuclear Magnetic Resonance (NMR) with a Bruker NMS Minispec and for paste samples an extraction with petroleum ether is used. Linoleic acid composition is measured by Gas Chromatography (GC), for all samples. These procedures are annexed at the end of this report. Then practical part is shown, followed by the results and their discussion. A total of 41 samples (36 calibration samples and 5 unknown samples) are scanned in the reflectance mode. Calibration equations for each parameter are developed using the regression method of PLS with internal cross validation. Finally conclusions are shown, with the appropriate reflections.