Determination of total trans fat content in Pakistani cereal-based foods by SB-HATR FT-IR spectroscopy coupled with partial least square regression

The purpose of this study was to determine the total trans fat content from the Fourier transform infrared (FT-IR) spectra of neat fats extracted from cereal-based foods using a single-bounce horizontal attenuated total reflectance (SB-HATR) sampling accessory. Partial least squares (PLS) regression was employed for the development of calibration models. A trans fat calibration curve was prepared by adding trielaidine gravimetrically to trans free canola oil. Regression of the FT-IR/PLS-predicted trans contents was 0.999 with a root-mean-square error of calibration (RMSEC) and root-mean-square error of prediction (RMSEP) 0.07% and 0.07%, respectively. Amongst the samples examined, the fat content was in the range of 8.2–27.2%, whilst most samples contained more than 10% trans fat; the highest trans fat content was 16.3%. The FT-IR results were comparable with gas chromatography (GC).