Non-destructive analysis of caviar compositions using low-field nuclear magnetic resonance technique

Abstract

Caviar is one of the most popular and expensive animal products in world trade. Water, fat and protein contents are key chemical compositions, and account for the majority of mass of caviar. In this study, the performance of a rapid, accurate analysis of water, fat and protein analysis was reported in caviar and raw sturgeon eggs through low-field 1H nuclear magnetic resonance (LF-NMR) relaxometry combined with partial least-squares regression (PLSR) models. For caviar, the correlation coefficients (Rcv 2) of water, fat and protein were 0.9930, 0.9698 and 0.9783 with root mean square error of cross-validation (RMSECV) of 0.0760, 0.0308 and 0.0566, respectively. For raw sturgeon eggs the Rcv 2 were 0.9932, 0.9592 and 0.9770, and the RMSECV were 0.1098, 0.0878 and 0.0917, respectively. Besides, a LF-NMR and principal component analysis (PCA) combined method also was developed to discriminate the caviar and raw sturgeon eggs based on the hydrogen protons originating from different environments with various salt concentrations.