Fat crystallisation in emulsions is a complex process. One of the important parameters is the solid fat content (SFC). Up to now, there is no standardised method to measure the SFC in emulsions, let alone to determine the SFC of the fat inside droplets, thus avoiding the signal of the aqueous phase. This work evaluates the capabilities of deconvolution of the free induction decay (FID)-Carr-Purcell-Meiboom-Gill (CPMG) signal of emulsions. Three models were evaluated. The first model was a combination of a Gaussian function and a bi-exponential function (GBE model). The second model combined a Gaussian function with multiple exponential functions (GME model). The last model contained multiple Gaussian functions and multiple exponential functions (MGME model). The latter two models used a simplified CONTIN analysis. Based on the analysis of the determination coefficient R2 , the calculated water content and the estimated SFC of nonemulsified two-phase systems, the GBE model was selected to analyse the FID-CPMG signal of emulsified systems. However, the results obtained with the other models did not differ substantially, and hence, they could be used to obtain a full relaxation time distribution. When the GBE model was applied on different emulsion systems, no significant differences in estimated SFC of the fat phase were found, thus indicating that the emulsion formulation (i.e. water-in-oil [W/O], oil-in-water [O/W] or water-in-oil-in-water [W/O/W]) only had a minor effect on the SFC in the systems considered here.
Keywords: 1H; NMR; SFC; TD-NMR; deconvolution; emulsions; lipids.
© 2022 John Wiley & Sons, Ltd.