Authentication of coriander oil and adulterant identification using electronic nose and spectroscopic techniques

Coriander oil possesses excellent antimicrobial, antioxidant, and anti-inflammatory properties, making it a valuable product for health care and food preservation. Due to its high value and desirable properties, this type of oil is often targeted for adulteration, either by blending it with less expensive oils or by completely substituting it. This study aims to authenticate coriander oil and distinguish the adulterants using non-invasive, green, fast and reliable methods. Performances of portable Raman and near infrared (NIR) spectrometers, as well as an electronic nose, were compared to distinguish between authentic and adulterated samples, using data driven soft independent class analogy (DD-SIMCA) and one class partial least squares (OC-PLS) models. DD-SIMCA achieved accuracies ranging from 87 % with NIR to 97 % with Raman spectroscopy, while OC-PLS obtained 69 % with NIR and up to 95 % with Raman. Moreover, a multiclass model was developed using only the adulterated samples to classify them into three types of adulterants: canola and soybean oils and palm olein. Partial Least Squares Discriminant (PLS-DA) successfully classified the type of adulterant, achieving 97 % accuracy with Raman spectroscopy. Raman demonstrated the highest performance; however, the custom-built, low-cost electronic nose also achieved high accuracy, with 94 % using DD-SIMCA and 96 % using PLS-DA, highlighting its potential as a rapid and effective screening tool. Therefore, the synergy of multivariate techniques and spectroscopic instruments was demonstrated by addressing both the verification of the authenticity of coriander oil and the detection of the type of adulterants.