Prediction of Slip Melting Point of Interesterified Fats by Means of FT-NIR Spectroscopy

Interesterified (IE) fats are used as hard fat replacers in a large range of commonly consumed foods including spreads, bakery and confectionary products. IE fats can provide desirable functional properties related to the product melting point, texture and shelf stability, while lowering the saturated fatty acid content. The exchange of the fatty acid groups in triacylglycerols alters the melting properties of the fat or fat blend, thus improving the functional properties. The chemical or enzymatic interesterification reaction can be monitored by analysing the slip melting point (SMP) of the IE fat. The aim of this work was the application of FT-NIR spectroscopy to the prediction of SMP of IE fats in order to provide producers a fast and reliable tool for the interesterification monitoring. Samples (75) of IE fats obtained by both chemical and enzymatic interesterification of tallow with different ratios of vegetable oils (corn, canola, and safflower oil) were analysed for SMP. The FT-NIR spectra (12500-3600 cm-1; res: 8 cm-1; scan: 64) were acquired both in transmission mode on solid samples (25\ub0C) and in transflectance mode on liquid samples (60\ub0C). Partial Least Squares regression (PLS) models for the prediction of SMP were calculated with FT-NIR data after reduction of the spectral range and application of different pre-treatments (smoothing, MSC, first and second derivatives). The best models were then recalculated using only the wavenumbers selected by the Martens uncertainty test in order to obtain simpler and more robust models. All models were validated by a cross-validation procedure, with 5 CV. The SMP ranged from 31.8 to 50.4\ub0C and the best PLS models provided a good prediction with both melted and solid sample spectra (R2CV, 0.78-0.88; RMSECV, 1.6-2.1\ub0C; 10 LVs). FT-NIR spectroscopy can be then considered a promising tool for an effective monitoring of fat interesterification in the production line