Prediction of fatty acid composition of sunflower seeds by near-infrared reflectance spectroscopy

Abstract

This study was performed in order to evaluate efficiency of near-infrared reflectance spectroscopy (NIRS) for the determination of fatty acid composition ratio of sunflower seeds and to compare performance of calibration methods. Calibration equations were developed using modified partial least squares (MPLS) and partial least squares (PLS) regression methods. Ninety-three sunflower seed varieties were from test field of East Mediterranean Agricultural Research Institute. In order to determine the reference fatty acid values needed to construct calibration in NIRS analysis, sunflower seed samples were analyzed by gas chromatography method. Coefficients of determination (R-2) in calibration were developed using MPLS and PLS as follows: for palmitic acid 0.706-0.664, for stearic acid 0.615-0.654, for oleic acid 0.996-0.994, for linoleic acid 0.995-0.994, for arachidic acid 0.768-0.643, for linolenic acid 0.818-0.763, for behenic acid 0.891-0.776, for eicosapentaenoic 0.933-0.892, for unsaturated fatty acid 0.837-0.890 and for saturated fatty acid 0.837-0.890 respectively. The results showed that NIRS was a reliable technique that can be used as a tool for rapid pre-screening of fatty acid composition of sunflower seeds.