Application of partial least square regression to differential scanning calorimetry data for fatty acid quantitation in olive oil

Abstract

A chemometric approach based on partial least (PLS) square methodology was applied to unfolded differential scanning calorimetry data obtained by 63 samples of different vegetable oils (58 extra virgin olive oils, one olive and one pomace olive oil, three seed oils) to evaluate fatty acid composition (palmitic, stearic, oleic and linoleic acids, saturated (SFA), mono (MUFA) and polysaturated (PUFA) percentages, oleic/linoleic and unsaturated/saturated ratios). All calibration models exhibited satisfactory figures of merit. Palmitic and oleic acids, as well as SFA showed very good correlation coefficients and low root mean square error values in both calibration and validation sets. Satisfactory results were also obtained for MUFA, PUFA, stearic and linoleic acids, O/L ratio in terms of percentage recoveries and relative standard deviations. No systematic and bias errors were detected in the prediction of validation samples. This novel approach could provide statistically similar results to those given by traditional official procedures, with the advantages of a very rapid and environmentally friendly methodology.