Non-destructive detection of heavy metals in vegetable oil based on nano-chemoselective response dye combined with near-infrared spectroscopy

Abstract Heavy metal concentrations are one problem bedeviling the market and consumption of edible oil. This study attempts to use near-infrared spectroscopy (NIRS) combined with chemoselective responsive dyes, as capture probes for the quantification of lead (Pb) and mercury (Hg) heavy metals in oils. Olfactory visualization system was used to screen chemoselective responsive dyes. The synthesized porous silica nanospheres (PSNs) were used to further optimize the color sensor and applied based on selected dyes. The spectral data were preprocessed by standard normal variation (SNV), which follows the application of chemometrics like partial least squares (PLS), ant colony optimization-PLS (ACO-PLS), synergy interval partial least squares (SiPLS), genetic algorithm-PLS (GA-PLS), competitive adaptive reweighted sampling-PLS (CARS-PLS) and partial least squares (PLS) were combined to construct the regression model. ACO-PLS achieved optimum result, with the Rp2 value of 0.9612 in the linear range of 0.001 ∼ 100ppm, and LOD of ≤ 1ppb recorded. Verified by the National Standard Detection Method, the effectiveness of this strategy has proven to be satisfactorily accurate. Therefore, the developed method could be used for non-destructive detection of lead and mercury in edible oil.