FTIR analysis for authentication and fatty acid determination of walnut oil adulterated with common vegetable oils
Georgiana Fediuc, Mircea Oroian
Abstract
: This study evaluated the ability of FTIR spectroscopy to authenticate walnut oil and detect adulteration with sunflower, rapeseed, and soybean oils at levels of 5–50%. A total of 196 samples (16 authentic and 180 adulterated) were analyzed using Attenuated Total Reflectance (ATR) sampling. The spectra exhibited characteristic triglyceride absorption peaks, with key differences observed in the 3100–2700 cm -1 and 1800–1000 cm -1 regions, though authentic and adulterated oils showed substantial spectral overlap. Chemometric analysis improved discrimination: Support Vector Machines (SVM) outperformed PLS-DA in classifying authentic versus adulterated oils, while Partial Least Squares Regression (PLS-R) accurately predicted adulteration levels and fatty acid composition (C16:0, C18:0, C18:1, C18:2, C18:3, C22:2, MUFA, PUFA). Pre-treatments optimized model performance, with normaliation effective for sunflower, de-trending for soybean, first derivative for rapeseed, and baseline correction for mixed adulterations (R 2 val = 0.656–0.982). Overall, FTIR spectroscopy combined with chemometric models offers a rapid, reliable, and non-destructive method for walnut oil authentication and fatty acid profiling.