Accelerated shelf life testing and spectroscopic modeling for rapid determination of free fatty acids in desiccated coconut
Description
Desiccated coconut is susceptible to lipid hydrolysis during storage, forming free fatty acids (FFA). The study aimed to predict the shelf life of desiccated coconut and develop near-infrared (NIR), Fourier-transform infrared (FT-IR), and Raman spectroscopic models for rapid determination of FFA content. Samples were stored at 25, 40, and 50 °C for 18 weeks, and FFA content and spectral data were collected at predetermined intervals. Shelf life was predicted using the Arrhenius model. Partial least squares (PLS) regression models were developed and validated for FFA determination. FFA formation followed zero-order kinetics, with increasing reaction rates at higher storage temperatures. The Arrhenius model yielded an overestimated shelf life of 6,712 days at 30 °C, substantially exceeding the typical industrial shelf life range. This may be due to the low FFA formation rate observed during the reaction's induction period. All calibration models yielded a coefficient of determination (R2) around 0.90. External validation confirmed that the models have satisfactory performance with low bias values, low root mean square error of prediction (RMSEP) (NIR = 0.0042, FT-IR = 0.0043, Raman = 0.0040), and ratio of performance deviation (RPD) values above two (NIR = 2.52, FT-IR = 2.19, Raman = 2.85). Raman spectroscopy showed the highest predictive performance among evaluated techniques. Overall, the findings indicate that the developed spectroscopic models provide a rapid, non-destructive approach for efficient quality evaluation of desiccated coconut, while highlighting that incorporating the induction period into kinetic modeling may further improve the reliability of shelf-life determination.
Files
Institutions
- University of Sri JayewardenepuraWestern Province, Colombo