ABSTRACT
Despite the advances in low-field nuclear magnetic resonance (NMR), there are limited spectroscopic applications for untargeted analysis and metabolomics. To evaluate its potential, we combined high-field and low-field NMR with chemometrics for the differentiation between virgin and refined coconut oil and for the detection of adulteration in blended samples. Although low-field NMR has less spectral resolution and sensitivity compared to high-field NMR, it was still able to achieve a differentiation between virgin and refined coconut oils, as well as between virgin coconut oil and blends, using principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and random forest techniques. These techniques were not able to distinguish between blends with different levels of adulteration; however, partial least squares regression (PLSR) enabled the quantification of adulteration levels for both NMR approaches. Given the significant benefits of low-field NMR, including economic and user-friendly analysis and fitting in an industrial environment, this study establishes the proof of concept for its utilization in the challenging scenario of coconut oil authentication. Also, this method has the potential to be used for other similar applications that involve untargeted analysis.