A novel approach for improved honey identification and scientific definition: a case of buckwheat honey

This study aimed to provide insights into a novel honey screening and authentication approach. Chemometrics associated with a GC/MS instrumentation was applied to verify a total of 13 honey samples having different floral and geographical origin with an aim to detect buckwheat honey and discriminate it from honeys of other floral sources. Non-polar and semi-polar compounds were first extracted and then detected and semi-quantified employing a GC/MS device working in a full scan mode. Fingerprinting signals covering peaks that elute in the first 20 min, comprising 2000 scans of semi-polar compounds, were used as input datasets for unsupervised and supervised chemometric tools. This approach demonstrated excellent classification performances of honey samples according to the declared floral source, regardless of their geographical origin. The obtained results were compared with the outputs of a common melissopalynological procedure. Although exhibiting a high variability in the pollen content, samples declared as buckwheat honey demonstrated similar chemical profiles. The proposed non-targeted and semi-quantitative method showed to be rapid, unbiased, independent of tedious qualitative and quantitative determinations of eluting compounds, thus exhibiting a strong potential to be incorporated into data fusion procedures for honey authentication. Harmonising chemical methodologies with certain indications from pollen analysis to form core-data systems would undoubtedly improve the authentication protocols of honey floral sources.