Non-destructive Detection of Sesame Oil Adulteration by Portable FTNIR, FT-MIR, and Raman Spectrometers Combined with Chemometrics

Abstract

Edible oils are often adulterated with fixed oils because of their high quality and price. Sesameoil is prone to adulteration due to its high commodity value and popularity. Therefore, a rapid, simple,and non-invasive method to detect adulteration in sesame oil is necessary for quality control purposes.Handheld and portable FT-NIR, FT-MIR, and Raman spectrometers are easy to operate, non-destructive,rapid, and easy to transport for in-situ assessments as well as being cheaper alternatives to traditionalinstruments. This study aimed to evaluate three different vibrational spectroscopic techniques indetecting sesame oil adulteration with sunflower and canola oil. Sesame oils were adulterated with fixedoils at different concentrations (0 – 25%) (w/w). Spectra were collected with portable devices andanalyzed using Soft Independent Modelling of Class Analogy (SIMCA) to generate a classification model toauthenticate pure sesame oil and Partial Least Squares Regression (PLSR) to predict the levels of theadulterant. For confirmation, the fatty acid profile of the oils was determined by gas chromatography(GC). In all three instruments, SIMCA provided distinct clusters for pure sesame oils and adulteratedsamples with interclass distance (ICD) over 3. Furthermore, FT-NIR and FT-MIR showed excellentperformance in predicting adulterant levels with rval>0.96. Specifically, the FT-MIR unit provided moreprecise classification and PLSR prediction models over FT-NIR and Raman units. Still, all the units can beused as an alternative method to traditional methods such as GC, GC-MS, etc. These units showed greatpotential for in-situ surveillance to detect sesame oil adulterations.