Determination of the geographic origin of 52 honey samples based on the assessment of anionic content profiling with a new algorithm using monolithic column-based micellar nano-liquid chromatography.

In the present study, a new micellar nano LC-UV was, for the first time, reported for the separation and determination of five anions (chloride, nitrite, bromide, sulfate and nitrate) in 52 honey samples. Based on this approach, a graphene oxide-based monolithic column was prepared and applied for the samples. Various amounts of hexadecyltrimethyl-ammonium bromide (HTAB) in the mobile phase were used in order to optimize the separation conditions. The baseline separation was achieved using mobile phase with 25/75% (v/v) ACN/10 mM phosphate buffer at pH 3.4, while the amount of HTAB was optimized as 0.22 mM in the mobile phase. The whole method was validated and it leads to high sensitivity. The LOD values were found in the range of 0.02-0.22 µg/kg, while LOQ values were found in the range of 0.06-0.18 µg/kg. The method allowed to achieve sensitivity analyses of anionic content in 52 honey samples. All data were evaluated using a new algorithm for geographic origin discrimination. K-nearest neighbor algorithm (K-NN), cubic support vector classifier (K-DVS), and K-Mean cluster analysis were used for geographic origin discrimination of honeys. The accuracy of the whole model was calculated as 94.4% with the K-DVS method. The samples from five provinces were classified 100% correctly, while two of them were classified with one misclassification, with an accuracy of 89.9% and 83.3%, respectively. PRACTICAL APPLICATION: The new platforms and advanced technologies are crucial for advanced food analysis. In this article, a novel methodology was attempted for the determination of geographic origin of 52 honey samples. In this sense, micellar nano LC technique with a homemade monolithic nano-column was, for the first time, applied for the anion analysis using a new algorithm.