Honey is a natural product that is highly appreciated for its nutritional and therapeutic properties, but is very susceptible to fraudulent practices. Its high economic value and complex composition make honey an attractive product for fraud., Fraudulent practices may involve adding adulterants, artificial feeding of the bees, or incorrect labelling of the botanical variety and geographical origin. In this context, there is a need for simple, fast, and reliable analytical methods for honey characterization and authentication.
A non-targeted cyclic voltammetry (CV) fingerprinting method using screen-printed (SPCE) electrodes and chemometric analysis to discriminate Spanish honey samples according to their principal typology (blossom or honeydew honey) and its specific botanical variety was developed and validated. One hundred and ten blossom and honeydew honey samples of different botanical varieties (blossom, eucalyptus, heather, lavender, rosemary, thyme, almond, chestnut, forest, holm oak, and mountain) were analyzed after a simple dissolution of 2 g of honey in 20 mL of water. CV measurements were initiated by immersing an SPCE into the honey extract solution while scanning the potential from -1.8 to 1.0 V for three consecutive cycles using a scan rate of 0.05 V/s. The second voltammetric cycle was selected as sample chemical descriptor for chemometrics. Exploratory principal component analysis showed clear discrimination among blossom and honeydew honey samples, except for blossom heather samples, which were grouped with the honeydew ones because of their similar physicochemical properties. Partial least squares discriminant analysis provided in general calibration and cross-validation sensitivity and specificity values higher than 0.9, and percentages of correct classification (PCC) ranging 80-92%. Good results were also achieved when addressing blossom and honeydew groups independently, with PCC values ranging from 70 to 100% and 73 to 93% for blossom and honeydew varieties, respectively, demonstrating the suitability of CV fingerprints as sample chemical descriptors to assess honey botanical variety.
