Comparative evaluation of low- and high-field ¹H NMR for vegetable oil authentication: classification,...

The authentication of vegetable oils is essential for consumer protection and regulatory compliance. This study compares low-field (LF, 100 MHz) and high-field (HF, 600 MHz) proton nuclear magnetic resonance (¹H NMR) spectroscopy to evaluate their performance in classification, compositional analysis (43 parameters including acidity, peroxide value, spectrophotometric indices, etc.), and fraud detection. It also highlights their practical advantages and limitations for routine quality control. LF-NMR demonstrated limited classification performance, correctly identifying only 53 % of extra virgin olive oil (EVOO) and 62 % of virgin olive oil (VOO) versus lampante olive oil (LOO), with overall reduced accuracy compared to HF-NMR. Its quantification capabilities (limited to major quality parameters such as fatty acid composition) was also lower, with R² values ranging from 0.48 to 0.94. Adulteration detection of EVOO with three lower quality vegetable oils revealed moderate deviations with a relative standard deviation (RSD) of 18.6 %. However reliable detection was only achieved above ≥15-20 % v/v, making it more suitable for preliminary screening rather than precise compositional analysis. In contrast, HF-NMR achieved 100 % classification accuracy for sunflower oil (SO), high-oleic SO (HOSO), and pomace olive oil (POO), and 80-83 % for EVOO, VOO, and LOO, with an overall accuracy of 90 %. Partial Least Squares Regression (PLS-R) models confirmed its superior quantification ability, with R² values above 0.95 for key parameters (e.g., oleic acid, acidity, peroxide value), including an R² of 0.98 and root mean square error of prediction (RMSEP) of 3.0 for oleic acid. HF-NMR also detected adulterants as low as 5.0 % v/v with RMSE of cross-validation (RMSE-CV) below 5.0 for all of them. This study delineates the trade-offs between analytical performance, accessibility, and cost-effectiveness of LF and HF-NMR. While HF-NMR remains the gold standard for compositional analysis, LF-NMR shows promise as a cost-effective screening tool.